Skip to content
Back to search

ai-engineering-from-scratch

Learn it. Build it. Ship it for others.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today

Check it deeper on your machine

It has changed since its scan on 11 Nov 2026.

As a Claude Code plugin, install it and then ask Claude Code:

/plugin marketplace add keithmackay/mackayi/plugin install scanskillsafety@mackayiDo a deep scan of https://github.com/rohitg00/ai-engineering-from-scratch with scanskillsafety

Or from a terminal, clone it and run it:

git clone https://github.com/keithmackay/scanskillsafety-skill ~/.claude/skills/scanskillsafetynode ~/.claude/skills/scanskillsafety/cli.cjs --deep 'https://github.com/rohitg00/ai-engineering-from-scratch'

It checks the exact commit: every file, its packages against OSV (api.osv.dev), how it compares with this page's scan (findsafeskills.com receives the repository's address), and, if you agree, Claude reading every file with your own Claude subscription. It never runs the code.

Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/rohitg00-ai-engineering-from-scratch/badge.svg)](/listings/rohitg00-ai-engineering-from-scratch)

HTML

<a href="/listings/rohitg00-ai-engineering-from-scratch"><img src="/api/listings/rohitg00-ai-engineering-from-scratch/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Three ways in. Pick one.

**Option A — learn in your terminal *(recommended)*.** After the Node.js, `npx`, host, and scope preflight above, install the learning skills into a compatible agent and let the course drive itself:

npx skills add rohitg00/ai-engineering-from-scratch

Use the host-specific invocation table above. The installed skills provide `start-learning`, `learn`, `course-guide`, and the focused `learn-mcp` and `learn-agent-skills` routes. Lesson prose can stream from this repository without a clone. A local clone is required for copied repository code commands and executable MCP or Agent Skills labs. Progress lives in `LEARNING.md`, `MCP-LEARNING.md`, or `AGENT-SKILLS-LEARNING.md` in your project, so every session can resume.

**Option B — read.** Open any completed lesson on [aiengineeringfromscratch.com](https://aiengineeringfromscratch.com) or expand a phase under [Contents](#contents). No setup, no cloning.

**Option C — clone and run.**

git clone https://github.com/rohitg00/ai-engineering-from-scratch.git
cd ai-engineering-from-scratch
python3 phases/01-math-foundations/01-linear-algebra-intuition/code/vectors.py

Cloning also auto-loads the learning skills in Claude Code, and gives every lesson's code to the `learn` tutor for real execution instead of read-along.

Prerequisites

- You can write code (any language; Python helps). - You want to understand how AI **actually works**, not just call APIs.

Prepare for Claude certifications

The [Claude Certification Academy](certifications/claude/README.md) is a free, open-source preparation program for all four official Claude certification tracks: Associate Foundations, Developer Foundations, Architect Foundations, and Architect Professional. Each route combines blueprint-mapped lessons, runnable labs, a diagnostic, capstone work, and a full-length original practice exam.

Use the [AI-native GitHub onboarding guide](certifications/claude/GETTING_STARTED.md) with Claude Code, Codex, ChatGPT, Cursor, or another agent. Run `claude-certification` in Codex, `/claude-certification` in Claude Code, or ask another host to use `claude-certification`. It chooses a track, creates a persistent route in `CLAUDE-CERTIFICATION.md`, teaches one step at a time, runs the real labs, and gives artifact-based feedback. The same curriculum remains available on the [certification website](https://aiengineeringfromscratch.com/certifications.html).

The academy is independent study material based on public exam objectives. It is not affiliated with Anthropic, does not reproduce live exam questions, and cannot guarantee a passing score.

Prepare for the MCP Associate (MCPA) certification

The [MCPA Certification Curriculum](certifications/mcpa/README.md) is a free, open-source preparation program for the Model Context Protocol Associate exam from the Agentic AI Foundation, delivered through Linux Foundation Training. Its 34 lessons teach the stateless 2026-07-28 protocol across the five exam domains: per-request `_meta` and `server/discover` in place of the old handshake, multi round-trip requests, subscriptions, caching, the tasks and MCP Apps extensions, OAuth authorization, and the registry and SDK tiers. Every lesson ships a runnable standard-library lab whose transcript is checked for the current wire shape, and the track adds a diagnostic, a capstone, and three full-length original practice exams whose question mix follows the published blueprint weights.

Use the [AI-native GitHub onboarding guide](certifications/mcpa/GETTING_STARTED.md) with Claude Code, Codex, ChatGPT, Cursor, or another agent. Run `mcpa-certification` in Codex, `/mcpa-certification` in Claude Code, or ask another host to use `mcpa-certification`. It creates a persistent route in `MCPA-CERTIFICATION.md`, teaches one step at a time, runs the real labs, and gives artifact-based feedback. The same curriculum is available on the

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 1,497 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. <p align="center">
  2. <picture>
  3. <source media="(prefers-color-scheme: dark)" srcset="assets/readme/header-dark.svg">
  4. <img src="assets/readme/header-light.svg" alt="AI Engineering from Scratch" width="840">
  5. </picture>
  6. </p>
  7. <p align="center">
  8. <a href="https://www.star-history.com/rohitg00/ai-engineering-from-scratch">
  9. <picture><source media="(prefers-color-scheme: dark)" srcset="https://api.star-history.com/badge?repo=rohitg00/ai-engineering-from-scratch&type=rank&theme=dark" /><source media="(prefers-color-scheme: light)" srcset="https://api.star-history.com/badge?repo=rohitg00/ai-engineering-from-scratch&type=rank" /><img alt="Star History Rank" src="https://api.star-history.com/badge?repo=rohitg00/ai-engineering-from-scratch&type=rank" /></picture> <picture><source media="(prefers-color-scheme: dark)" srcset="https://api.star-history.com/badge?repo=rohitg00/ai-engineering-from-scratch&type=trending&theme=dark" /><source media="(prefers-color-scheme: light)" srcset="https://api.star-history.com/badge?repo=rohitg00/ai-engineering-from-scratch&type=trending" /><img alt="GitHub Trending Repository of the Day" src="https://api.star-history.com/badge?repo=rohitg00/ai-engineering-from-scratch&type=trending" /></picture>
  10. </a>
  11. </p>
  12. Implement model internals, retrieval pipelines, and agent runtimes. Test them, inspect failures, and keep the code and evaluation results.
  13. **[Start learning](https://aiengineeringfromscratch.com/learn/setup-and-tooling/dev-environment)** · **[Choose a path](#learning-routes)** · **[Try a lab](#interactive-lab)** · **[Build a project](#project-challenges)** · **[Browse the curriculum](#contents)**
  14. Free, open source, MIT. Learn on the website, with a coding agent, or by running local code.
  15. > 542 lessons. 20 phases. Python, TypeScript, Rust, Julia, Go.
  16. Turn your coding agent into a tutor with one command:
  17. ```bash
  18. npx skills add rohitg00/ai-engineering-from-scratch
  19. ```
  20. <p align="center">
  21. <a href="LICENSE"><img src="https://img.shields.io/badge/license-MIT-1a1a1a?style=flat-square&labelColor=fafaf5" alt="MIT License"></a>
  22. <a href="ROADMAP.md"><img src="https://img.shields.io/badge/lessons-542-3553ff?style=flat-square&labelColor=fafaf5" alt="542 lessons"></a>
  23. <a href="#contents"><img src="https://img.shields.io/badge/phases-20-3553ff?style=flat-square&labelColor=fafaf5" alt="20 phases"></a>
  24. <a href="https://github.com/rohitg00/ai-engineering-from-scratch/stargazers"><img src="https://img.shields.io/github/stars/rohitg00/ai-engineering-from-scratch?style=flat-square&labelColor=fafaf5&color=3553ff" alt="GitHub stars"></a>
  25. <a href="https://aiengineeringfromscratch.com"><img src="https://img.shields.io/badge/web-aiengineeringfromscratch.com-3553ff?style=flat-square&labelColor=fafaf5" alt="Website"></a>
  26. </p>
  27. <details>
  28. <summary>Read in your language</summary>
  29. <p align="center">
  30. <a href="README.md">🇬🇧 English</a> ·
  31. <a href="i18n/zh/README.md">🇨🇳 简体中文</a> ·
  32. <a href="i18n/zh-TW/README.md">🇹🇼 繁體中文(台灣)</a> ·
  33. <a href="i18n/ja/README.md">🇯🇵 日本語</a> ·
  34. <a href="i18n/ko/README.md">🇰🇷 한국어</a> ·
  35. <a href="i18n/pt/README.md">🇵🇹 Português</a> ·
  36. <a href="i18n/pt-BR/README.md">🇧🇷 Português (Brasil)</a> ·
  37. <a href="i18n/es/README.md">🇪🇸 Español</a> ·
  38. <a href="i18n/de/README.md">🇩🇪 Deutsch</a> ·
  39. <a href="i18n/fr/README.md">🇫🇷 Français</a> ·
  40. <a href="i18n/it/README.md">🇮🇹 Italiano</a> ·
  41. <a href="i18n/nl/README.md">🇳🇱 Nederlands</a> ·
  42. <a href="i18n/pl/README.md">🇵🇱 Polski</a> ·
  43. <a href="i18n/cs/README.md">🇨🇿 Čeština</a> ·
  44. <a href="i18n/ro/README.md">🇷🇴 Română</a> ·
  45. <a href="i18n/hu/README.md">🇭🇺 Magyar</a> ·
  46. <a href="i18n/el/README.md">🇬🇷 Ελληνικά</a> ·
  47. <a href="i18n/sv/README.md">🇸🇪 Svenska</a> ·
  48. <a href="i18n/da/README.md">🇩🇰 Dansk</a> ·
  49. <a href="i18n/no/README.md">🇳🇴 Norsk</a> ·
  50. <a href="i18n/fi/README.md">🇫🇮 Suomi</a> ·
  51. <a href="i18n/ru/README.md">🇷🇺 Русский</a> ·
  52. <a href="i18n/uk/README.md">🇺🇦 Українська</a> ·
  53. <a href="i18n/tr/README.md">🇹🇷 Türkçe</a> ·
  54. <a href="i18n/he/README.md">🇮🇱 עברית</a> ·
  55. <a href="i18n/ar/README.md">🇸🇦 العربية</a> ·
  56. <a href="i18n/fa/README.md">🇮🇷 فارسی</a> ·
  57. <a href="i18n/hi/README.md">🇮🇳 हिन्दी</a> ·
  58. <a href="i18n/bn/README.md">🇧🇩 বাংলা</a> ·
  59. <a href="i18n/ur/README.md">🇵🇰 اردو</a> ·
  60. <a href="i18n/th/README.md">🇹🇭 ไทย</a> ·
  61. <a href="i18n/vi/README.md">🇻🇳 Tiếng Việt</a> ·
  62. <a href="i18n/id/README.md">🇮🇩 Bahasa Indonesia</a> ·
  63. <a href="i18n/tl/README.md">🇵🇭 Tagalog</a>
  64. </p>
  65. </details>
  66. ### Sponsors
  67. <p align="center">
  68. <a href="https://serpapi.com/ai-engineering-from-scratch"><picture><source media="(min-width: 768px)" srcset="assets/sponsors/serpapi-banner-compact.png" width="48%"><img src="assets/sponsors/serpapi-banner-compact.png" alt="SerpApi. Web Search API for your AI apps. Available in Markdown and JSON for any integration." width="440"></picture></a>
  69. </p>
  70. <p align="center">
  71. <sub><span>Your support keeps every lesson free and open source.</span> <a href="#supporters">See all supporters</a> · <a href="SPONSORS.md">Become a sponsor</a></sub>
  72. </p>
  73. ## Why build it from scratch
  74. A framework call hides the part that breaks. When a loss turns into NaN or an agent repeats the same tool call, the person who wrote the small version knows where to look. Most lessons have you write that small version first and run it, then show where production tools do the same work.
  75. <p align="center">
  76. <picture>
  77. <source media="(max-width: 600px) and (prefers-color-scheme: dark)" srcset="assets/readme/lesson-loop-mobile-dark.svg">
  78. <source media="(max-width: 600px)" srcset="assets/readme/lesson-loop-mobile-light.svg">
  79. <source media="(prefers-color-scheme: dark)" srcset="assets/readme/lesson-loop-dark.svg">
  80. <img src="assets/readme/lesson-loop-light.svg" alt="One lesson in four steps: the problem, build it from scratch, see where production tools do the same work, and ship an artifact you keep. The example is the agent loop lesson in Phase 14." width="840">
  81. </picture>
  82. </p>
  83. <details>
  84. <summary>Text version</summary>
  85. - **01 · THE PROBLEM**: An LLM returns text. It cannot act.
  86. - **02 · BUILD IT**: A ReAct loop in stdlib Python.
  87. - **03 · USE IT**: Where agent SDKs run the same loop.
  88. - **04 · SHIP IT**: skill-agent-loop.md, ready for your agent.
  89. </details>
  90. <a id="see-what-you-will-build-and-keep"></a>
  91. <a id="learning-routes"></a>
  92. ## Learning paths
  93. | Route | Starting lesson |
  94. |---|---|
  95. | Model foundations | [Setup and tooling](https://aiengineeringfromscratch.com/learn/setup-and-tooling/dev-environment) |
  96. | LLM systems | [Prompt engineering](https://aiengineeringfromscratch.com/learn/llm-engineering/prompt-engineering) |
  97. | Agents and delivery | [The agent loop](https://aiengineeringfromscratch.com/learn/agent-engineering/the-agent-loop) |
  98. [Compare career paths](https://aiengineeringfromscratch.com/learning-paths.html) · [Prerequisites and study time](#study-guide)
  99. <a id="interactive-lab"></a>
  100. ### Gradient descent
  101. Twenty starting points follow gradient descent on a quadratic loss. The graph shows their positions and mean loss after each update.
  102. <p align="center">
  103. <a href="https://aiengineeringfromscratch.com/learn/math-foundations/optimization#loss-landscape-visualization">
  104. <picture>
  105. <source media="(max-width: 600px) and (prefers-color-scheme: dark) and (prefers-reduced-motion: reduce)" srcset="assets/readme/101-gradient-mobile-dark.png">
  106. <source media="(max-width: 600px) and (prefers-reduced-motion: reduce)" srcset="assets/readme/101-gradient-mobile-light.png">
  107. <source media="(prefers-color-scheme: dark) and (prefers-reduced-motion: reduce)" srcset="assets/readme/101-gradient-dark.png">
  108. <source media="(prefers-reduced-motion: reduce)" srcset="assets/readme/101-gradient-light.png">
  109. <source media="(max-width: 600px) and (prefers-color-scheme: dark)" srcset="assets/readme/101-gradient-mobile-dark.gif">
  110. <source media="(max-width: 600px)" srcset="assets/readme/101-gradient-mobile-light.gif">
  111. <source media="(prefers-color-scheme: dark)" srcset="assets/readme/101-gradient-dark.gif">
  112. <img src="assets/readme/101-gradient-light.gif" alt="Gradient descent moves scattered starting points toward the loss minimum. The mean loss decreases with each update." width="840">
  113. </picture>
  114. </a>
  115. </p>
  116. [Adjust the learning rate in the lesson](https://aiengineeringfromscratch.com/learn/math-foundations/optimization#loss-landscape-visualization) · [Compare GD, momentum, and Adam in code](phases/01-math-foundations/08-optimization/code/optimizers.py)
  117. <a id="project-challenges"></a>
  118. ### Projects
  119. Three projects with staged starters, reference implementations, and local graders. Run commands from the repository root after [setup](#local-setup). Starters fail until you implement the stages.
  120. <details>
  121. <summary><strong>01 · Retrieval Evaluation Lab</strong> · Python · Ranking metrics and regression checks</summary>
  122. A candidate improves mean NDCG while one query ranks its most relevant evidence lower. Build a query-by-query comparison that reports the regression and can fail a release check.
  123. Use Python 3.10+. Review [RAG](phases/11-llm-engineering/06-rag/docs/en.md) and [model evaluation](phases/02-ml-fundamentals/09-model-evaluation/docs/en.md). Implement ranking validation, precision and recall, rank-sensitive metrics, then system comparison.
  124. ```bash
  125. python3 scripts/project_test.py retrieval-evaluation-lab \
  126. --init learning-artifacts/retrieval-evaluation-lab
  127. python3 scripts/project_test.py retrieval-evaluation-lab \
  128. --stage 1 --path learning-artifacts/retrieval-evaluation-lab --strict
  129. python3 scripts/project_test.py retrieval-evaluation-lab \
  130. --all --path learning-artifacts/retrieval-evaluation-lab --strict
  131. ```
  132. **Keep:** a reproducible comparison with per-query deltas and the judgments used to score them. Metrics describe those judgments; they do not establish answer correctness.
  133. [Start the project](https://aiengineeringfromscratch.com/project.html?id=retrieval-evaluation-lab) · [Inspect the reference](projects/retrieval-evaluation-lab/solution/) · [Run on your own inputs](projects/retrieval-evaluation-lab/README.md#run-with-your-own-inputs)
  134. </details>
  135. <details>
  136. <summary><strong>02 · Agent Trace Debugger</strong> · TypeScript · Trace parsing and timing</summary>
  137. A supplied trace still takes 100 ms, but total token usage rises by 200 and one span starts failing. Separate overlapping child work from parent execution time, then produce a report that exposes the change.
  138. Use Node.js 22.18+ and Python 3 for the grader. Implement JSONL parsing, parent validation, interval arithmetic, then an inspectable timeline.
  139. ```bash
  140. python3 scripts/project_test.py agent-trace-debugger \
  141. --init learning-artifacts/agent-trace-debugger
  142. python3 scripts/project_test.py agent-trace-debugger \
  143. --stage 1 --path learning-artifacts/agent-trace-debugger --strict
  144. python3 scripts/project_test.py agent-trace-debugger \
  145. --all --path learning-artifacts/agent-trace-debugger --strict
  146. ```
  147. **Keep:** the input trace, an HTML timeline, and a JSON regression report. Preserve exclusive per-span token counts so parent and child usage are not counted twice.
  148. [Start the project](https://aiengineeringfromscratch.com/project.html?id=agent-trace-debugger) · [Inspect the reference](projects/agent-trace-debugger/solution/) · [Explore timing interactively](https://aiengineeringfromscratch.com/project.html?id=agent-trace-debugger&stage=03-timing)
  149. </details>
  150. <details>
  151. <summary><strong>03 · Tool Call Firewall</strong> · Rust · Role checks and approval receipts</summary>
  152. A write changes after review, or an approval is replayed. Validate the call envelope, check the caller's role and path, then consume an approval bound to the exact request and content.
  153. Use Rust and Python 3.10+. Review [tool schema design](phases/13-tools-and-protocols/05-tool-schema-design/docs/en.md) and [security boundaries](phases/17-infrastructure-and-production/25-security-secrets-audit/docs/en.md). The caller application supplies identity; the model proposes an operation.
  154. ```bash
  155. python3 scripts/project_test.py tool-call-firewall \
  156. --init learning-artifacts/tool-call-firewall
  157. python3 scripts/project_test.py tool-call-firewall \
  158. --stage 1 --path learning-artifacts/tool-call-firewall --strict
  159. python3 scripts/project_test.py tool-call-firewall \
  160. --all --path learning-artifacts/tool-call-firewall --strict
  161. ```
  162. **Keep:** an audit receipt showing the requested operation and the policy decision. Approvals are single-use within one invocation; this project does not provide persistent authorization or an operating-system sandbox.
  163. [Start the project](https://aiengineeringfromscratch.com/project.html?id=tool-call-firewall) · [Inspect the reference](projects/tool-call-firewall/solution/) · [Explore approval boundaries](https://aiengineeringfromscratch.com/project.html?id=tool-call-firewall&stage=03-consume-a-request-bound-approval-once)
  164. </details>
  165. [Browse all projects](https://aiengineeringfromscratch.com/projects.html) · [Career practice guide](learning-paths/CAREER-PRACTICE.md)
  166. ## Choose how to learn
  167. ### On the website
  168. Open any completed lesson on [aiengineeringfromscratch.com](https://aiengineeringfromscratch.com) or expand a phase under [Contents](#contents). No setup, no cloning.
  169. ### With an AI tutor
  170. If Node.js, `npx`, and a skill-capable coding agent are already installed, your coding agent can become your tutor. A repository clone is not needed to install or read the tutor. Runnable focused-path labs need `python3`. Agent Skills host labs also need a selected host and a writable user or project skill scope.
  171. ```bash
  172. npx skills add rohitg00/ai-engineering-from-scratch
  173. ```
  174. Choose the host and scope when the installer asks. Use `start-learning` in Codex, `/start-learning` in Claude Code, or ask your host to use the skill by name.
  175. <details>
  176. <summary>Tutor setup and host commands</summary>
  177. Check the local requirements first:
  178. ```bash
  179. node --version
  180. npx --version
  181. python3 --version
  182. ```
  183. `skills` writes to the host and scope selected during installation, such as
  184. `.claude/skills/`, `.cursor/skills/`, `.codex/skills/`, or another supported
  185. skills folder. Verify that the selected host discovers that exact destination.
  186. Invocation syntax belongs to the host, not to the portable `SKILL.md` format:
  187. | Host | Start the course | Start Model Context Protocol (MCP) | Start Agent Skills | Run a phase quiz |
  188. |---|---|---|---|---|
  189. | Codex | `start-learning`, or choose it from `/skills` | `learn-mcp`, or choose it from `/skills` | `learn-agent-skills`, or choose it from `/skills` | `check-understanding 13`, or choose it from `/skills` |
  190. | Claude Code | `/start-learning` | `/learn-mcp` | `/learn-agent-skills` | `/check-understanding 13` |
  191. | Other compatible hosts | `Use start-learning to begin the course.` | `Use learn-mcp to start the Model Context Protocol (MCP) path.` | `Use learn-agent-skills to start the Agent Skills Engineering path.` | `Use check-understanding to quiz me on Phase 13.` |
  192. A ten-question placement quiz maps what you already know to a starting phase and
  193. saves a personalized study plan to `LEARNING.md`. From there, the `learn` skill
  194. teaches one lesson per session: concept, math, code, quiz. It streams lessons
  195. straight from this repo, and the `course-guide` skill jumps you to the exact
  196. lesson that covers anything you are stuck on. In Codex, invoke these skills with
  197. `learn` and `course-guide`; in Claude Code, use `/learn` and `/course-guide`;
  198. in other compatible hosts, ask to use the skill by name.
  199. Only want Model Context Protocol (MCP)? Use the MCP invocation for your host. It creates
  200. `MCP-LEARNING.md` and follows one 17-lesson route through stateless
  201. requests, transports, bidirectional work, security, reliability, registry
  202. governance, and conformance evidence. The exact order and checkpoints live in
  203. the [Model Context Protocol (MCP) manifest](learning-paths/model-context-protocol.json).
  204. Only want Agent Skills? Use the Agent Skills invocation for your host. It
  205. creates `AGENT-SKILLS-LEARNING.md` and follows one coherent five-lesson route:
  206. contract, discovery, invocation, sandbox boundaries, then release evals and
  207. real-host portability. Start on the web with the
  208. [Agent Skills path](https://aiengineeringfromscratch.com/learn/tools-and-protocols/skills-and-agent-sdks?learningPath=agent-skills).
  209. The installer lists the hosts it can configure and asks where to install. If
  210. you do not have Node.js, `npx`, `python3`, a supported host, or a writable
  211. scope yet, use the website or read `docs/en.md` manually. That path teaches the
  212. concepts, but real-host discovery, invocation, script, and uninstall evidence
  213. remains pending until the preflight is available. Read the lessons at
  214. [aiengineeringfromscratch.com](https://aiengineeringfromscratch.com).
  215. ### The learning skills
  216. | Skill | What it does |
  217. |---|---|
  218. | [`start-learning`](skills/start-learning/SKILL.md) | One-time onboarding: why you're learning, placement quiz, personalized plan saved to `LEARNING.md`. |
  219. | [`learn`](skills/learn/SKILL.md) | The tutor loop. Warm-up recall, then the next lesson taught interactively, then its quiz; records progress and a review queue. |
  220. | [`course-guide`](skills/course-guide/SKILL.md) | Topic router. "Where do I learn attention?" or "my loss is NaN" → the exact lessons, with links. |
  221. | [`learn-mcp`](skills/learn-mcp/SKILL.md) | Focused Model Context Protocol (MCP) tutor. Creates `MCP-LEARNING.md`, follows the 17-lesson manifest, and records wire, security, reliability, and conformance evidence. |
  222. | [`learn-agent-skills`](skills/learn-agent-skills/SKILL.md) | Focused Agent Skills tutor. Creates `AGENT-SKILLS-LEARNING.md`, teaches lessons 22, 24, 25, 26, and 27, and records real-host evidence. |
  223. | [`learn-reverse-engineering`](skills/learn-reverse-engineering/SKILL.md) | Focused Reverse Engineering tutor. Creates `REVERSE-ENGINEERING-LEARNING.md`, teaches Phase 13 lessons 32 to 45 on programs you build yourself, and records lab evidence. |
  224. | [`claude-certification`](skills/claude-certification/SKILL.md) | Certification tutor. Chooses CCAO-F, CCDV-F, CCAR-F, or CCAR-P; teaches each lesson; runs labs; reviews artifacts; administers diagnostics and mocks; saves progress. |
  225. | [`mcpa-certification`](skills/mcpa-certification/SKILL.md) | MCPA tutor. Follows the 34-lesson `mcpa-f` route on the 2026-07-28 protocol; teaches each lesson; runs labs and the wire checker; administers the diagnostic and three mocks; saves progress. |
  226. | [`find-your-level`](skills/find-your-level/SKILL.md) | Ten-question placement quiz. Maps your knowledge to a starting phase and produces a personalized path with hour estimates. |
  227. | [`check-understanding <phase>`](skills/check-understanding/SKILL.md) | Per-phase quiz, eight questions, with feedback and specific lessons to review. Use the Codex, Claude Code, or natural-language form in the invocation table above. |
  228. </details>
  229. <a id="local-setup"></a>
  230. ### Run local code
  231. ```bash
  232. git clone https://github.com/rohitg00/ai-engineering-from-scratch.git
  233. cd ai-engineering-from-scratch
  234. python3 phases/00-setup-and-tooling/01-dev-environment/code/verify.py --route beginner
  235. python3 phases/01-math-foundations/01-linear-algebra-intuition/code/vectors.py
  236. ```
  237. The preflight separates requirements needed now from tools needed later. Every
  238. required failure includes the detected reason and a corrective command. The
  239. `vectors.py` command is a dependency-free lesson and ends by showing that a matrix
  240. times a vector is the operation inside a neural network layer. Save that
  241. terminal output as your first evidence.
  242. <details>
  243. <summary>Use every lesson the same way</summary>
  244. ### Use every lesson the same way
  245. 1. **Read** `docs/en.md` and explain the core idea in your own words.
  246. 2. **Type and build** the important code instead of treating the code block as decoration.
  247. 3. **Run** the lesson command from the repository root, the directory containing `README.md` and `phases/`.
  248. 4. **Keep evidence**: the command, working directory, exit code, meaningful output, and the artifact you changed or produced.
  249. 5. **Continue** only when you can explain the output and make one small change without guessing.
  250. Commands in lesson pages are paths from the repository root unless the lesson
  251. explicitly says to change directories. If a lesson offers several languages,
  252. run the implementation for the language you are learning.
  253. </details>
  254. <a id="study-guide"></a>
  255. ## Choose a learning path
  256. You do not need to scan 542 lessons before beginning. Pick one goal. Each link
  257. opens the same curriculum on GitHub or the website, and both versions use the
  258. same lesson code.
  259. | Your goal | Learn on GitHub | Learn on the website |
  260. |---|---|---|
  261. | I am new and want the complete foundation | [Phase 0: Setup and Tooling](phases/00-setup-and-tooling/) | [Dev Environment](https://aiengineeringfromscratch.com/learn/setup-and-tooling/dev-environment) |
  262. | I know Python and want math plus ML foundations | [Phase 1: Math Foundations](phases/01-math-foundations/) | [Linear Algebra Intuition](https://aiengineeringfromscratch.com/learn/math-foundations/linear-algebra-intuition) |
  263. | I want to build production LLM applications | [Phase 11: LLM Engineering](phases/11-llm-engineering/) | [Prompt Engineering](https://aiengineeringfromscratch.com/learn/llm-engineering/prompt-engineering) |
  264. | I want to build agents | [Phase 14: Agent Engineering](phases/14-agent-engineering/) | [The Agent Loop](https://aiengineeringfromscratch.com/learn/agent-engineering/the-agent-loop) |
  265. | I want to use coding agents on real repositories | [Agent-Assisted Engineering path](learning-paths/using-coding-agents.json) | [Agent-Assisted Engineering](https://aiengineeringfromscratch.com/learn/agent-engineering/agent-workbench-why-models-fail?learningPath=using-coding-agents) |
  266. | I want to shape the right build before implementation | [Product Judgment and Delivery path](learning-paths/shaping-the-build.json) | [Product Judgment and Delivery](https://aiengineeringfromscratch.com/learn/agent-engineering/outcomes-before-output?learningPath=shaping-the-build) |
  267. Not sure where you fit? Use the [`start-learning` placement tutor](skills/start-learning/SKILL.md)
  268. or the [website prerequisites guide](https://aiengineeringfromscratch.com/prereqs.html).
  269. Compare four core domains and six career routes in the [AI Engineering Learning Paths](https://aiengineeringfromscratch.com/learning-paths.html).
  270. <details>
  271. <summary>Focused MCP, Agent Skills, and Reverse Engineering paths</summary>
  272. | Your goal | Learn on GitHub | Learn on the website |
  273. |---|---|---|
  274. | I want to build with Model Context Protocol (MCP) | [Model Context Protocol (MCP) route](phases/13-tools-and-protocols/README.md#model-context-protocol-mcp-path) | [Model Context Protocol (MCP) path](https://aiengineeringfromscratch.com/learn/tools-and-protocols/mcp-fundamentals?learningPath=model-context-protocol) |
  275. | I want to write and ship Agent Skills | [Focused Agent Skills route](phases/13-tools-and-protocols/README.md#agent-skills-fast-path) | [Agent Skills path](https://aiengineeringfromscratch.com/learn/tools-and-protocols/skills-and-agent-sdks?learningPath=agent-skills) |
  276. | I want to read compiled programs | [Reverse Engineering route](phases/13-tools-and-protocols/README.md#reverse-engineering-path) | [Reverse Engineering path](https://aiengineeringfromscratch.com/learn/tools-and-protocols/program-anatomy-elf-from-scratch?learningPath=reverse-engineering) |
  277. </details>
  278. <details>
  279. <summary>Prerequisites and study time</summary>
  280. ### Prerequisites
  281. - You can write code (any language; Python helps).
  282. - You want to understand how AI **actually works**, not just call APIs.
  283. ## Where to start
  284. | Background | Start at | Estimated time |
  285. |---|---|---|
  286. | New to programming and AI | Phase 0 — Setup | ~306 hours |
  287. | Know Python, new to ML | Phase 1 — Math Foundations | ~270 hours |
  288. | Know ML, new to deep learning | Phase 3 — Deep Learning Core | ~200 hours |
  289. | Know deep learning, want LLMs and agents | Phase 10 — LLMs from Scratch | ~100 hours |
  290. | Senior engineer, only want agent engineering | Phase 14 — Agent Engineering | ~60 hours |
  291. | Only want to build production MCP systems | [Model Context Protocol (MCP) path](learning-paths/model-context-protocol.json) | ~23 hours 15 min |
  292. | Only want to build production Agent Skills | [Agent Skills Engineering path](learning-paths/agent-skills.json) | ~9.5 hours |
  293. | Only want to learn reverse engineering | [Reverse Engineering path](learning-paths/reverse-engineering.json) | ~32.5 hours |
  294. </details>
  295. ## The shape of the curriculum
  296. Twenty phases stack on top of each other. Math is the floor. Agents and production are the roof.
  297. Skip ahead if you already know the lower layers, but don't skip and then wonder why something at
  298. the top is breaking.
  299. <p align="center">
  300. <picture>
  301. <source media="(max-width: 600px) and (prefers-color-scheme: dark)" srcset="assets/readme/curriculum-map-mobile-dark.svg">
  302. <source media="(max-width: 600px)" srcset="assets/readme/curriculum-map-mobile-light.svg">
  303. <source media="(prefers-color-scheme: dark)" srcset="assets/readme/curriculum-map-dark.svg">
  304. <img src="assets/readme/curriculum-map-light.svg" alt="The twenty phases in six volumes, from foundations at the bottom to production at the top. Volume 1, Foundations, phases 0 to 2: vectors, gradients, classic models. Volume 2, Deep Learning, phases 3, 4 and 6: from a perceptron to vision and speech. Volume 3, Language, phases 5 and 7: attention and the transformer, by hand. Volume 4, Large Language Models, phases 8 to 11: pre-train a mini GPT, align it, ship RAG. Volume 5, Agents, phases 12 to 16: the agent loop, tools, MCP servers, swarms. Volume 6, Production, phases 17 to 19: serving, observability, safety, capstones." width="840">
  305. </picture>
  306. </p>
  307. <details>
  308. <summary>Phase dependency graph</summary>
  309. ```mermaid
  310. %%{init: {'theme':'base','themeVariables':{'primaryColor':'#fafaf5','primaryTextColor':'#1a1a1a','primaryBorderColor':'#3553ff','lineColor':'#3553ff','fontFamily':'JetBrains Mono','fontSize':'12px'}}}%%
  311. flowchart TB
  312. P0["Phase 0 — Setup &amp; Tooling"] --> P1["Phase 1 — Math Foundations"]
  313. P1 --> P2["Phase 2 — ML Fundamentals"]
  314. P2 --> P3["Phase 3 — Deep Learning Core"]
  315. P3 --> P4["Phase 4 — Vision"]
  316. P3 --> P5["Phase 5 — NLP"]
  317. P3 --> P6["Phase 6 — Speech &amp; Audio"]
  318. P3 --> P9["Phase 9 — RL"]
  319. P5 --> P7["Phase 7 — Transformers"]
  320. P7 --> P8["Phase 8 — GenAI"]
  321. P7 --> P10["Phase 10 — LLMs from Scratch"]
  322. P10 --> P11["Phase 11 — LLM Engineering"]
  323. P10 --> P12["Phase 12 — Multimodal"]
  324. P11 --> P13["Phase 13 — Tools &amp; Protocols"]
  325. P13 --> P14["Phase 14 — Agent Engineering"]
  326. P14 --> P15["Phase 15 — Autonomous Systems"]
  327. P15 --> P16["Phase 16 — Multi-Agent &amp; Swarms"]
  328. P14 --> P17["Phase 17 — Infrastructure &amp; Production"]
  329. P15 --> P18["Phase 18 — Ethics &amp; Alignment"]
  330. P16 --> P19["Phase 19 — Capstone Projects"]
  331. P17 --> P19
  332. P18 --> P19
  333. classDef phase fill:#fafaf5,stroke:#3553ff,color:#1a1a1a
  334. class P0,P1,P2,P3,P4,P5,P6,P7,P8,P9,P10,P11,P12,P13,P14,P15,P16,P17,P18,P19 phase
  335. ```
  336. </details>
  337. <a id="contents"></a>
  338. ## Contents
  339. Twenty phases. Click any phase to expand its lesson list.
  340. <a id="phase-0"></a>
  341. ### Phase 0: Setup & Tooling `13 lessons`
  342. > Get your environment ready for everything that follows.
  343. | # | Lesson | Type | Lang |
  344. |:---:|--------|:----:|------|
  345. | 01 | [Dev Environment](phases/00-setup-and-tooling/01-dev-environment/) | Build | Python, TypeScript, Rust |
  346. | 02 | [Git & Collaboration](phases/00-setup-and-tooling/02-git-and-collaboration/) | Learn | — |
  347. | 03 | [GPU Setup & Cloud](phases/00-setup-and-tooling/03-gpu-setup-and-cloud/) | Build | Python |
  348. | 04 | [APIs & Keys](phases/00-setup-and-tooling/04-apis-and-keys/) | Build | Python, TypeScript |
  349. | 05 | [Jupyter Notebooks](phases/00-setup-and-tooling/05-jupyter-notebooks/) | Build | Python |
  350. | 06 | [Python Environments](phases/00-setup-and-tooling/06-python-environments/) | Build | Shell |
  351. | 07 | [Docker for AI](phases/00-setup-and-tooling/07-docker-for-ai/) | Build | Docker |
  352. | 08 | [Editor Setup](phases/00-setup-and-tooling/08-editor-setup/) | Build | — |
  353. | 09 | [Data Management](phases/00-setup-and-tooling/09-data-management/) | Build | Python |
  354. | 10 | [Terminal & Shell](phases/00-setup-and-tooling/10-terminal-and-shell/) | Learn | — |
  355. | 11 | [Linux for AI](phases/00-setup-and-tooling/11-linux-for-ai/) | Learn | — |
  356. | 12 | [Debugging & Profiling](phases/00-setup-and-tooling/12-debugging-and-profiling/) | Build | Python |
  357. | 13 | [Python for AI Engineering](phases/00-setup-and-tooling/13-python-for-ai-engineering/) | Build | Python, TypeScript |
  358. <details id="phase-1">
  359. <summary><b>Phase 1 — Math Foundations</b> &nbsp;<code>22 lessons</code>&nbsp; <em>The intuition behind every AI algorithm, through code.</em></summary>
  360. <br/>
  361. | # | Lesson | Type | Lang |
  362. |:---:|--------|:----:|------|
  363. | 01 | [Linear Algebra Intuition](phases/01-math-foundations/01-linear-algebra-intuition/) | Learn | Python, Julia |
  364. | 02 | [Vectors, Matrices & Operations](phases/01-math-foundations/02-vectors-matrices-operations/) | Build | Python, Julia |
  365. | 03 | [Matrix Transformations & Eigenvalues](phases/01-math-foundations/03-matrix-transformations/) | Build | Python, Julia |
  366. | 04 | [Calculus for ML: Derivatives & Gradients](phases/01-math-foundations/04-calculus-for-ml/) | Learn | Python, Julia |
  367. | 05 | [Chain Rule & Automatic Differentiation](phases/01-math-foundations/05-chain-rule-and-autodiff/) | Build | Python, Julia |
  368. | 06 | [Probability & Distributions](phases/01-math-foundations/06-probability-and-distributions/) | Learn | Python, Julia |
  369. | 07 | [Bayes' Theorem & Statistical Thinking](phases/01-math-foundations/07-bayes-theorem/) | Build | Python, TypeScript |
  370. | 08 | [Optimization: Gradient Descent Family](phases/01-math-foundations/08-optimization/) | Build | Python, Julia |
  371. | 09 | [Information Theory: Entropy, KL Divergence](phases/01-math-foundations/09-information-theory/) | Learn | Python, TypeScript |
  372. | 10 | [Dimensionality Reduction: PCA, t-SNE, UMAP](phases/01-math-foundations/10-dimensionality-reduction/) | Build | Python, TypeScript, Julia |
  373. | 11 | [Singular Value Decomposition](phases/01-math-foundations/11-singular-value-decomposition/) | Build | Python, Julia |
  374. | 12 | [Tensor Operations](phases/01-math-foundations/12-tensor-operations/) | Build | Python, TypeScript |
  375. | 13 | [Numerical Stability](phases/01-math-foundations/13-numerical-stability/) | Build | Python, TypeScript, Julia |
  376. | 14 | [Norms & Distances](phases/01-math-foundations/14-norms-and-distances/) | Build | Python, TypeScript |
  377. | 15 | [Statistics for ML](phases/01-math-foundations/15-statistics-for-ml/) | Build | Python, TypeScript |
  378. | 16 | [Sampling Methods](phases/01-math-foundations/16-sampling-methods/) | Build | Python, TypeScript, Julia |
  379. | 17 | [Linear Systems](phases/01-math-foundations/17-linear-systems/) | Build | Python, TypeScript, Julia |
  380. | 18 | [Convex Optimization](phases/01-math-foundations/18-convex-optimization/) | Build | Python, TypeScript, Julia |
  381. | 19 | [Complex Numbers for AI](phases/01-math-foundations/19-complex-numbers/) | Learn | Python, TypeScript |
  382. | 20 | [The Fourier Transform](phases/01-math-foundations/20-fourier-transform/) | Build | Python, TypeScript, Julia |
  383. | 21 | [Graph Theory for ML](phases/01-math-foundations/21-graph-theory/) | Build | Python, TypeScript |
  384. | 22 | [Stochastic Processes](phases/01-math-foundations/22-stochastic-processes/) | Learn | Python, Julia |
  385. </details>
  386. <details id="phase-2">
  387. <summary><b>Phase 2 — ML Fundamentals</b> &nbsp;<code>21 lessons</code>&nbsp; <em>Classical ML — still the backbone of most production AI.</em></summary>
  388. <br/>
  389. | # | Lesson | Type | Lang |
  390. |:---:|--------|:----:|------|
  391. | 01 | [What Is Machine Learning](phases/02-ml-fundamentals/01-what-is-machine-learning/) | Learn | Python |
  392. | 02 | [Linear Regression from Scratch](phases/02-ml-fundamentals/02-linear-regression/) | Build | Python, Rust, Julia |
  393. | 03 | [Logistic Regression & Classification](phases/02-ml-fundamentals/03-logistic-regression/) | Build | Python, Rust, Julia |
  394. | 04 | [Decision Trees & Random Forests](phases/02-ml-fundamentals/04-decision-trees/) | Build | Python, TypeScript |
  395. | 05 | [Support Vector Machines](phases/02-ml-fundamentals/05-support-vector-machines/) | Build | Python, Julia |
  396. | 06 | [KNN & Distance Metrics](phases/02-ml-fundamentals/06-knn-and-distances/) | Build | Python, TypeScript |
  397. | 07 | [Unsupervised Learning: K-Means, DBSCAN](phases/02-ml-fundamentals/07-unsupervised-learning/) | Build | Python, TypeScript |
  398. | 08 | [Feature Engineering & Selection](phases/02-ml-fundamentals/08-feature-engineering/) | Build | Python, TypeScript |
  399. | 09 | [Model Evaluation: Metrics, Cross-Validation](phases/02-ml-fundamentals/09-model-evaluation/) | Build | Python, Julia |
  400. | 10 | [Bias, Variance & the Learning Curve](phases/02-ml-fundamentals/10-bias-variance/) | Learn | Python |
  401. | 11 | [Ensemble Methods: Boosting, Bagging, Stacking](phases/02-ml-fundamentals/11-ensemble-methods/) | Build | Python |
  402. | 12 | [Hyperparameter Tuning](phases/02-ml-fundamentals/12-hyperparameter-tuning/) | Build | Python |
  403. | 13 | [ML Pipelines & Experiment Tracking](phases/02-ml-fundamentals/13-ml-pipelines/) | Build | Python |
  404. | 14 | [Naive Bayes](phases/02-ml-fundamentals/14-naive-bayes/) | Build | Python |
  405. | 15 | [Time Series Fundamentals](phases/02-ml-fundamentals/15-time-series/) | Build | Python |
  406. | 16 | [Anomaly Detection](phases/02-ml-fundamentals/16-anomaly-detection/) | Build | Python |
  407. | 17 | [Handling Imbalanced Data](phases/02-ml-fundamentals/17-imbalanced-data/) | Build | Python |
  408. | 18 | [Feature Selection](phases/02-ml-fundamentals/18-feature-selection/) | Build | Python |
  409. | 19 | [ML System Design — From Problem to Deployed Model](phases/02-ml-fundamentals/19-ml-system-design/) | Build | Python, TypeScript |
  410. | 20 | [Monitoring Models in Production — Data Drift and Concept Drift](phases/02-ml-fundamentals/20-model-monitoring-and-drift/) | Build | Python, TypeScript |
  411. | 21 | [Feature Stores and Training-Serving Skew](phases/02-ml-fundamentals/21-feature-stores-and-training-serving-skew/) | Build | Python, TypeScript |
  412. </details>
  413. <details id="phase-3">
  414. <summary><b>Phase 3 — Deep Learning Core</b> &nbsp;<code>13 lessons</code>&nbsp; <em>Neural networks from first principles. No frameworks until you build one.</em></summary>
  415. <br/>
  416. | # | Lesson | Type | Lang |
  417. |:---:|--------|:----:|------|
  418. | 01 | [The Perceptron: Where It All Started](phases/03-deep-learning-core/01-the-perceptron/) | Build | Python, TypeScript, Rust, Julia |
  419. | 02 | [Multi-Layer Networks & Forward Pass](phases/03-deep-learning-core/02-multi-layer-networks/) | Build | Python, TypeScript, Julia |
  420. | 03 | [Backpropagation from Scratch](phases/03-deep-learning-core/03-backpropagation/) | Build | Python, TypeScript, Rust, Julia |
  421. | 04 | [Activation Functions: ReLU, Sigmoid, GELU & Why](phases/03-deep-learning-core/04-activation-functions/) | Build | Python, TypeScript, Julia |
  422. | 05 | [Loss Functions: MSE, Cross-Entropy, Contrastive](phases/03-deep-learning-core/05-loss-functions/) | Build | Python, TypeScript, Julia |
  423. | 06 | [Optimizers: SGD, Momentum, Adam, AdamW](phases/03-deep-learning-core/06-optimizers/) | Build | Python, TypeScript, Julia |
  424. | 07 | [Regularization: Dropout, Weight Decay, BatchNorm](phases/03-deep-learning-core/07-regularization/) | Build | Python, TypeScript, Julia |
  425. | 08 | [Weight Initialization & Training Stability](phases/03-deep-learning-core/08-weight-initialization/) | Build | Python, TypeScript, Julia |
  426. | 09 | [Learning Rate Schedules & Warmup](phases/03-deep-learning-core/09-learning-rate-schedules/) | Build | Python, TypeScript, Julia |
  427. | 10 | [Build Your Own Mini Framework](phases/03-deep-learning-core/10-mini-framework/) | Build | Python, TypeScript, Julia |
  428. | 11 | [Introduction to PyTorch](phases/03-deep-learning-core/11-intro-to-pytorch/) | Build | Python |
  429. | 12 | [Introduction to JAX](phases/03-deep-learning-core/12-intro-to-jax/) | Build | Python |
  430. | 13 | [Debugging Neural Networks](phases/03-deep-learning-core/13-debugging-neural-networks/) | Build | Python |
  431. </details>
  432. <details id="phase-4">
  433. <summary><b>Phase 4 — Computer Vision</b> &nbsp;<code>28 lessons</code>&nbsp; <em>From pixels to understanding — image, video, 3D, VLMs, and world models.</em></summary>
  434. <br/>
  435. | # | Lesson | Type | Lang |
  436. |:---:|--------|:----:|------|
  437. | 01 | [Image Fundamentals: Pixels, Channels, Color Spaces](phases/04-computer-vision/01-image-fundamentals/) | Learn | Python, Julia |
  438. | 02 | [Convolutions from Scratch](phases/04-computer-vision/02-convolutions-from-scratch/) | Build | Python, Julia |
  439. | 03 | [CNNs: LeNet to ResNet](phases/04-computer-vision/03-cnns-lenet-to-resnet/) | Build | Python |
  440. | 04 | [Image Classification](phases/04-computer-vision/04-image-classification/) | Build | Python |
  441. | 05 | [Transfer Learning & Fine-Tuning](phases/04-computer-vision/05-transfer-learning/) | Build | Python |
  442. | 06 | [Object Detection — YOLO from Scratch](phases/04-computer-vision/06-object-detection-yolo/) | Build | Python |
  443. | 07 | [Semantic Segmentation — U-Net](phases/04-computer-vision/07-semantic-segmentation-unet/) | Build | Python |
  444. | 08 | [Instance Segmentation — Mask R-CNN](phases/04-computer-vision/08-instance-segmentation-mask-rcnn/) | Build | Python |
  445. | 09 | [Image Generation — GANs](phases/04-computer-vision/09-image-generation-gans/) | Build | Python |
  446. | 10 | [Image Generation — Diffusion Models](phases/04-computer-vision/10-image-generation-diffusion/) | Build | Python |
  447. | 11 | [Stable Diffusion — Architecture & Fine-Tuning](phases/04-computer-vision/11-stable-diffusion/) | Build | Python |
  448. | 12 | [Video Understanding — Temporal Modeling](phases/04-computer-vision/12-video-understanding/) | Build | Python |
  449. | 13 | [3D Vision: Point Clouds, NeRFs](phases/04-computer-vision/13-3d-vision-nerf/) | Build | Python |
  450. | 14 | [Vision Transformers (ViT)](phases/04-computer-vision/14-vision-transformers/) | Build | Python |
  451. | 15 | [Real-Time Vision: Edge Deployment](phases/04-computer-vision/15-real-time-edge/) | Build | Python, Rust |
  452. | 16 | [Build a Complete Vision Pipeline](phases/04-computer-vision/16-vision-pipeline-capstone/) | Build | Python |
  453. | 17 | [Self-Supervised Vision — SimCLR, DINO, MAE](phases/04-computer-vision/17-self-supervised-vision/) | Build | Python |
  454. | 18 | [Open-Vocabulary Vision — CLIP](phases/04-computer-vision/18-open-vocab-clip/) | Build | Python |
  455. | 19 | [OCR & Document Understanding](phases/04-computer-vision/19-ocr-document-understanding/) | Build | Python |
  456. | 20 | [Image Retrieval & Metric Learning](phases/04-computer-vision/20-image-retrieval-metric/) | Build | Python |
  457. | 21 | [Keypoint Detection & Pose Estimation](phases/04-computer-vision/21-keypoint-pose/) | Build | Python |
  458. | 22 | [3D Gaussian Splatting from Scratch](phases/04-computer-vision/22-3d-gaussian-splatting/) | Build | Python |
  459. | 23 | [Diffusion Transformers & Rectified Flow](phases/04-computer-vision/23-diffusion-transformers-rectified-flow/) | Build | Python |
  460. | 24 | [SAM 3 & Open-Vocabulary Segmentation](phases/04-computer-vision/24-sam3-open-vocab-segmentation/) | Build | Python |
  461. | 25 | [Vision-Language Models (ViT-MLP-LLM)](phases/04-computer-vision/25-vision-language-models/) | Build | Python |
  462. | 26 | [Monocular Depth & Geometry Estimation](phases/04-computer-vision/26-monocular-depth/) | Build | Python |
  463. | 27 | [Multi-Object Tracking & Video Memory](phases/04-computer-vision/27-multi-object-tracking/) | Build | Python, Rust |
  464. | 28 | [World Models & Video Diffusion](phases/04-computer-vision/28-world-models-video-diffusion/) | Build | Python |
  465. </details>
  466. <details id="phase-5">
  467. <summary><b>Phase 5 — NLP: Foundations to Advanced</b> &nbsp;<code>29 lessons</code>&nbsp; <em>Language is the interface to intelligence.</em></summary>
  468. <br/>
  469. | # | Lesson | Type | Lang |
  470. |:---:|--------|:----:|------|
  471. | 01 | [Text Processing: Tokenization, Stemming, Lemmatization](phases/05-nlp-foundations-to-advanced/01-text-processing/) | Build | Python, TypeScript |
  472. | 02 | [Bag of Words, TF-IDF & Text Representation](phases/05-nlp-foundations-to-advanced/02-bag-of-words-tfidf/) | Build | Python, TypeScript |
  473. | 03 | [Word Embeddings: Word2Vec from Scratch](phases/05-nlp-foundations-to-advanced/03-word-embeddings-word2vec/) | Build | Python, Julia |
  474. | 04 | [GloVe, FastText & Subword Embeddings](phases/05-nlp-foundations-to-advanced/04-glove-fasttext-subword/) | Build | Python, TypeScript |
  475. | 05 | [Sentiment Analysis](phases/05-nlp-foundations-to-advanced/05-sentiment-analysis/) | Build | Python, TypeScript |
  476. | 06 | [Named Entity Recognition (NER)](phases/05-nlp-foundations-to-advanced/06-named-entity-recognition/) | Build | Python, TypeScript |
  477. | 07 | [POS Tagging & Syntactic Parsing](phases/05-nlp-foundations-to-advanced/07-pos-tagging-parsing/) | Build | Python, TypeScript |
  478. | 08 | [Text Classification — CNNs & RNNs for Text](phases/05-nlp-foundations-to-advanced/08-cnns-rnns-for-text/) | Build | Python, TypeScript |
  479. | 09 | [Sequence-to-Sequence Models](phases/05-nlp-foundations-to-advanced/09-sequence-to-sequence/) | Build | Python, TypeScript |
  480. | 10 | [Attention Mechanism — The Breakthrough](phases/05-nlp-foundations-to-advanced/10-attention-mechanism/) | Build | Python, TypeScript |
  481. | 11 | [Machine Translation](phases/05-nlp-foundations-to-advanced/11-machine-translation/) | Build | Python, TypeScript |
  482. | 12 | [Text Summarization](phases/05-nlp-foundations-to-advanced/12-text-summarization/) | Build | Python, TypeScript |
  483. | 13 | [Question Answering Systems](phases/05-nlp-foundations-to-advanced/13-question-answering/) | Build | Python, TypeScript |
  484. | 14 | [Information Retrieval & Search](phases/05-nlp-foundations-to-advanced/14-information-retrieval-search/) | Build | Python, TypeScript, Rust |
  485. | 15 | [Topic Modeling: LDA, BERTopic](phases/05-nlp-foundations-to-advanced/15-topic-modeling/) | Build | Python, TypeScript, Julia |
  486. | 16 | [Text Generation](phases/05-nlp-foundations-to-advanced/16-text-generation-pre-transformer/) | Build | Python, TypeScript |
  487. | 17 | [Chatbots: Rule-Based to Neural](phases/05-nlp-foundations-to-advanced/17-chatbots-rule-to-neural/) | Build | Python, TypeScript |
  488. | 18 | [Multilingual NLP](phases/05-nlp-foundations-to-advanced/18-multilingual-nlp/) | Build | Python, TypeScript |
  489. | 19 | [Subword Tokenization: BPE, WordPiece, Unigram, SentencePiece](phases/05-nlp-foundations-to-advanced/19-subword-tokenization/) | Learn | Python, TypeScript, Rust |
  490. | 20 | [Structured Outputs & Constrained Decoding](phases/05-nlp-foundations-to-advanced/20-structured-outputs-constrained-decoding/) | Build | Python, TypeScript, Rust |
  491. | 21 | [NLI & Textual Entailment](phases/05-nlp-foundations-to-advanced/21-nli-textual-entailment/) | Learn | Python, TypeScript |
  492. | 22 | [Embedding Models Deep Dive](phases/05-nlp-foundations-to-advanced/22-embedding-models-deep-dive/) | Learn | Python, TypeScript |
  493. | 23 | [Chunking Strategies for RAG](phases/05-nlp-foundations-to-advanced/23-chunking-strategies-rag/) | Build | Python, TypeScript |
  494. | 24 | [Coreference Resolution](phases/05-nlp-foundations-to-advanced/24-coreference-resolution/) | Learn | Python, TypeScript |
  495. | 25 | [Entity Linking & Disambiguation](phases/05-nlp-foundations-to-advanced/25-entity-linking/) | Build | Python, TypeScript |
  496. | 26 | [Relation Extraction & Knowledge Graph Construction](phases/05-nlp-foundations-to-advanced/26-relation-extraction-kg/) | Build | Python, TypeScript |
  497. | 27 | [LLM Evaluation: RAGAS, DeepEval, G-Eval](phases/05-nlp-foundations-to-advanced/27-llm-evaluation-frameworks/) | Build | Python, TypeScript |
  498. | 28 | [Long-Context Evaluation: NIAH, RULER, LongBench, MRCR](phases/05-nlp-foundations-to-advanced/28-long-context-evaluation/) | Learn | Python, TypeScript |
  499. | 29 | [Dialogue State Tracking](phases/05-nlp-foundations-to-advanced/29-dialogue-state-tracking/) | Build | Python, TypeScript |
  500. </details>
  501. <details id="phase-6">
  502. <summary><b>Phase 6 — Speech & Audio</b> &nbsp;<code>17 lessons</code>&nbsp; <em>Hear, understand, speak.</em></summary>
  503. <br/>
  504. | # | Lesson | Type | Lang |
  505. |:---:|--------|:----:|------|
  506. | 01 | [Audio Fundamentals: Waveforms, Sampling, FFT](phases/06-speech-and-audio/01-audio-fundamentals) | Learn | Python, TypeScript, Julia |
  507. | 02 | [Spectrograms, Mel Scale & Audio Features](phases/06-speech-and-audio/02-spectrograms-mel-features) | Build | Python, TypeScript, Julia |
  508. | 03 | [Audio Classification](phases/06-speech-and-audio/03-audio-classification) | Build | Python, TypeScript |
  509. | 04 | [Speech Recognition (ASR)](phases/06-speech-and-audio/04-speech-recognition-asr) | Build | Python, TypeScript |
  510. | 05 | [Whisper: Architecture & Fine-Tuning](phases/06-speech-and-audio/05-whisper-architecture-finetuning) | Build | Python, TypeScript |
  511. | 06 | [Speaker Recognition & Verification](phases/06-speech-and-audio/06-speaker-recognition-verification) | Build | Python, TypeScript |
  512. | 07 | [Text-to-Speech (TTS)](phases/06-speech-and-audio/07-text-to-speech) | Build | Python, TypeScript |
  513. | 08 | [Voice Cloning & Voice Conversion](phases/06-speech-and-audio/08-voice-cloning-conversion) | Build | Python, TypeScript |
  514. | 09 | [Music Generation](phases/06-speech-and-audio/09-music-generation) | Build | Python, TypeScript |
  515. | 10 | [Audio-Language Models](phases/06-speech-and-audio/10-audio-language-models) | Build | Python, TypeScript |
  516. | 11 | [Real-Time Audio Processing](phases/06-speech-and-audio/11-real-time-audio-processing) | Build | Python, Rust |
  517. | 12 | [Build a Voice Assistant Pipeline](phases/06-speech-and-audio/12-voice-assistant-pipeline) | Build | Python, TypeScript |
  518. | 13 | [Neural Audio Codecs — EnCodec, SNAC, Mimi, DAC](phases/06-speech-and-audio/13-neural-audio-codecs) | Learn | Python, TypeScript |
  519. | 14 | [Voice Activity Detection & Turn-Taking](phases/06-speech-and-audio/14-voice-activity-detection-turn-taking) | Build | Python, TypeScript, Rust |
  520. | 15 | [Streaming Speech-to-Speech — Moshi, Hibiki](phases/06-speech-and-audio/15-streaming-speech-to-speech-moshi-hibiki) | Learn | Python, TypeScript |
  521. | 16 | [Voice Anti-Spoofing & Audio Watermarking](phases/06-speech-and-audio/16-anti-spoofing-audio-watermarking) | Build | Python, TypeScript |
  522. | 17 | [Audio Evaluation — WER, MOS, MMAU, Leaderboards](phases/06-speech-and-audio/17-audio-evaluation-metrics) | Learn | Python, TypeScript |
  523. </details>
  524. <details id="phase-7">
  525. <summary><b>Phase 7 — Transformers Deep Dive</b> &nbsp;<code>16 lessons</code>&nbsp; <em>The architecture that changed everything.</em></summary>
  526. <br/>
  527. | # | Lesson | Type | Lang |
  528. |:---:|--------|:----:|------|
  529. | 01 | [Why Transformers: The Problems with RNNs](phases/07-transformers-deep-dive/01-why-transformers/) | Learn | Python, Julia |
  530. | 02 | [Self-Attention from Scratch](phases/07-transformers-deep-dive/02-self-attention-from-scratch/) | Build | Python, Rust, Julia |
  531. | 03 | [Multi-Head Attention](phases/07-transformers-deep-dive/03-multi-head-attention/) | Build | Python, Rust |
  532. | 04 | [Positional Encoding: Sinusoidal, RoPE, ALiBi](phases/07-transformers-deep-dive/04-positional-encoding/) | Build | Python, Rust, Julia |
  533. | 05 | [The Full Transformer: Encoder + Decoder](phases/07-transformers-deep-dive/05-full-transformer/) | Build | Python, Julia |
  534. | 06 | [BERT — Masked Language Modeling](phases/07-transformers-deep-dive/06-bert-masked-language-modeling/) | Build | Python, Rust |
  535. | 07 | [GPT — Causal Language Modeling](phases/07-transformers-deep-dive/07-gpt-causal-language-modeling/) | Build | Python, Rust |
  536. | 08 | [T5, BART — Encoder-Decoder Models](phases/07-transformers-deep-dive/08-t5-bart-encoder-decoder/) | Learn | Python, Rust |
  537. | 09 | [Vision Transformers (ViT)](phases/07-transformers-deep-dive/09-vision-transformers/) | Build | Python, Rust |
  538. | 10 | [Audio Transformers — Whisper Architecture](phases/07-transformers-deep-dive/10-audio-transformers-whisper/) | Learn | Python, Rust |
  539. | 11 | [Mixture of Experts (MoE)](phases/07-transformers-deep-dive/11-mixture-of-experts/) | Build | Python, Rust |
  540. | 12 | [KV Cache, Flash Attention & Inference Optimization](phases/07-transformers-deep-dive/12-kv-cache-flash-attention/) | Build | Python, Rust |
  541. | 13 | [Scaling Laws](phases/07-transformers-deep-dive/13-scaling-laws/) | Learn | Python, Rust |
  542. | 14 | [Build a Transformer from Scratch](phases/07-transformers-deep-dive/14-build-a-transformer-capstone/) | Build | Python |
  543. | 15 | [Attention Variants — Sliding Window, Sparse, Differential](phases/07-transformers-deep-dive/15-attention-variants/) | Build | Python, Rust |
  544. | 16 | [Speculative Decoding — Draft, Verify, Repeat](phases/07-transformers-deep-dive/16-speculative-decoding/) | Build | Python, Rust |
  545. </details>
  546. <details id="phase-8">
  547. <summary><b>Phase 8 — Generative AI</b> &nbsp;<code>15 lessons</code>&nbsp; <em>Create images, video, audio, 3D, and more.</em></summary>
  548. <br/>
  549. | # | Lesson | Type | Lang |
  550. |:---:|--------|:----:|------|
  551. | 01 | [Generative Models: Taxonomy & History](phases/08-generative-ai/01-generative-models-taxonomy-history/) | Learn | Python, TypeScript |
  552. | 02 | [Autoencoders & VAE](phases/08-generative-ai/02-autoencoders-vae/) | Build | Python, TypeScript |
  553. | 03 | [GANs: Generator vs Discriminator](phases/08-generative-ai/03-gans-generator-discriminator/) | Build | Python, TypeScript |
  554. | 04 | [Conditional GANs & Pix2Pix](phases/08-generative-ai/04-conditional-gans-pix2pix/) | Build | Python, TypeScript |
  555. | 05 | [StyleGAN](phases/08-generative-ai/05-stylegan/) | Build | Python, TypeScript |
  556. | 06 | [Diffusion Models — DDPM from Scratch](phases/08-generative-ai/06-diffusion-ddpm-from-scratch/) | Build | Python, TypeScript |
  557. | 07 | [Latent Diffusion & Stable Diffusion](phases/08-generative-ai/07-latent-diffusion-stable-diffusion/) | Build | Python, TypeScript |
  558. | 08 | [ControlNet, LoRA & Conditioning](phases/08-generative-ai/08-controlnet-lora-conditioning/) | Build | Python, TypeScript |
  559. | 09 | [Inpainting, Outpainting & Editing](phases/08-generative-ai/09-inpainting-outpainting-editing/) | Build | Python, TypeScript |
  560. | 10 | [Video Generation](phases/08-generative-ai/10-video-generation/) | Build | Python, TypeScript |
  561. | 11 | [Audio Generation](phases/08-generative-ai/11-audio-generation/) | Build | Python, TypeScript |
  562. | 12 | [3D Generation](phases/08-generative-ai/12-3d-generation/) | Build | Python, TypeScript |
  563. | 13 | [Flow Matching & Rectified Flows](phases/08-generative-ai/13-flow-matching-rectified-flows/) | Build | Python, TypeScript, Julia |
  564. | 14 | [Evaluation: FID, CLIP Score](phases/08-generative-ai/14-evaluation-fid-clip-score/) | Build | Python, TypeScript |
  565. | 19 | [Visual Autoregressive Modeling (VAR): Next-Scale Prediction](phases/08-generative-ai/19-visual-autoregressive-var/) | Build | Python, Julia |
  566. </details>
  567. <details id="phase-9">
  568. <summary><b>Phase 9 — Reinforcement Learning</b> &nbsp;<code>12 lessons</code>&nbsp; <em>The foundation of RLHF and game-playing AI.</em></summary>
  569. <br/>
  570. | # | Lesson | Type | Lang |
  571. |:---:|--------|:----:|------|
  572. | 01 | [MDPs, States, Actions & Rewards](phases/09-reinforcement-learning/01-mdps-states-actions-rewards/) | Learn | Python, TypeScript |
  573. | 02 | [Dynamic Programming](phases/09-reinforcement-learning/02-dynamic-programming/) | Build | Python, TypeScript |
  574. | 03 | [Monte Carlo Methods](phases/09-reinforcement-learning/03-monte-carlo-methods/) | Build | Python, TypeScript |
  575. | 04 | [Q-Learning, SARSA](phases/09-reinforcement-learning/04-q-learning-sarsa/) | Build | Python, TypeScript |
  576. | 05 | [Deep Q-Networks (DQN)](phases/09-reinforcement-learning/05-dqn/) | Build | Python, TypeScript |
  577. | 06 | [Policy Gradients — REINFORCE](phases/09-reinforcement-learning/06-policy-gradients-reinforce/) | Build | Python, TypeScript |
  578. | 07 | [Actor-Critic — A2C, A3C](phases/09-reinforcement-learning/07-actor-critic-a2c-a3c/) | Build | Python, TypeScript |
  579. | 08 | [PPO](phases/09-reinforcement-learning/08-ppo/) | Build | Python, TypeScript |
  580. | 09 | [Reward Modeling & RLHF](phases/09-reinforcement-learning/09-reward-modeling-rlhf/) | Build | Python, TypeScript |
  581. | 10 | [Multi-Agent RL](phases/09-reinforcement-learning/10-multi-agent-rl/) | Build | Python, TypeScript |
  582. | 11 | [Sim-to-Real Transfer](phases/09-reinforcement-learning/11-sim-to-real-transfer/) | Build | Python, TypeScript |
  583. | 12 | [RL for Games](phases/09-reinforcement-learning/12-rl-for-games/) | Build | Python, TypeScript |
  584. </details>
  585. <details id="phase-10">
  586. <summary><b>Phase 10 — LLMs from Scratch</b> &nbsp;<code>24 lessons</code>&nbsp; <em>Build, train, and understand large language models.</em></summary>
  587. <br/>
  588. | # | Lesson | Type | Lang |
  589. |:---:|--------|:----:|------|
  590. | 01 | [Tokenizers: BPE, WordPiece, SentencePiece](phases/10-llms-from-scratch/01-tokenizers/) | Build | Python, Rust |
  591. | 02 | [Building a Tokenizer from Scratch](phases/10-llms-from-scratch/02-building-a-tokenizer/) | Build | Python, TypeScript |
  592. | 03 | [Data Pipelines for Pre-Training](phases/10-llms-from-scratch/03-data-pipelines/) | Build | Python, Rust |
  593. | 04 | [Pre-Training a Mini GPT (124M)](phases/10-llms-from-scratch/04-pre-training-mini-gpt/) | Build | Python, Rust |
  594. | 05 | [Distributed Training, FSDP, DeepSpeed](phases/10-llms-from-scratch/05-scaling-distributed/) | Build | Python |
  595. | 06 | [Instruction Tuning — SFT](phases/10-llms-from-scratch/06-instruction-tuning-sft/) | Build | Python |
  596. | 07 | [RLHF — Reward Model + PPO](phases/10-llms-from-scratch/07-rlhf/) | Build | Python |
  597. | 08 | [DPO — Direct Preference Optimization](phases/10-llms-from-scratch/08-dpo/) | Build | Python |
  598. | 09 | [Constitutional AI & Self-Improvement](phases/10-llms-from-scratch/09-constitutional-ai-self-improvement/) | Build | Python |
  599. | 10 | [Evaluation — Benchmarks, Evals](phases/10-llms-from-scratch/10-evaluation/) | Build | Python |
  600. | 11 | [Quantization: INT8, GPTQ, AWQ, GGUF](phases/10-llms-from-scratch/11-quantization/) | Build | Python, Rust |
  601. | 12 | [Inference Optimization](phases/10-llms-from-scratch/12-inference-optimization/) | Build | Python, Rust |
  602. | 13 | [Building a Complete LLM Pipeline](phases/10-llms-from-scratch/13-building-complete-llm-pipeline/) | Build | Python, Rust |
  603. | 14 | [Open Models: Architecture Walkthroughs](phases/10-llms-from-scratch/14-open-models-architecture-walkthroughs/) | Learn | Python, Rust |
  604. | 15 | [Speculative Decoding and EAGLE-3](phases/10-llms-from-scratch/15-speculative-decoding-eagle3/) | Build | Python, Rust |
  605. | 16 | [Differential Attention (V2)](phases/10-llms-from-scratch/16-differential-attention-v2/) | Build | Python, Rust |
  606. | 17 | [Native Sparse Attention (DeepSeek NSA)](phases/10-llms-from-scratch/17-native-sparse-attention/) | Build | Python, Rust |
  607. | 18 | [Multi-Token Prediction (MTP)](phases/10-llms-from-scratch/18-multi-token-prediction/) | Build | Python, Rust |
  608. | 19 | [DualPipe Parallelism](phases/10-llms-from-scratch/19-dualpipe-parallelism/) | Learn | Python, Rust |
  609. | 20 | [DeepSeek-V3 Architecture Walkthrough](phases/10-llms-from-scratch/20-deepseek-v3-walkthrough/) | Learn | Python, Rust |
  610. | 21 | [Jamba — Hybrid SSM-Transformer](phases/10-llms-from-scratch/21-jamba-hybrid-ssm-transformer/) | Learn | Python, Rust |
  611. | 22 | [Async and Hogwild! Inference](phases/10-llms-from-scratch/22-async-hogwild-inference/) | Build | Python, Rust |
  612. | 25 | [Speculative Decoding and EAGLE](phases/10-llms-from-scratch/25-speculative-decoding/) | Build | Python |
  613. | 34 | [Gradient Checkpointing and Activation Recomputation](phases/10-llms-from-scratch/34-gradient-checkpointing/) | Build | Python |
  614. </details>
  615. <details id="phase-11">
  616. <summary><b>Phase 11 — LLM Engineering</b> &nbsp;<code>17 lessons</code>&nbsp; <em>Put LLMs to work in production.</em></summary>
  617. <br/>
  618. | # | Lesson | Type | Lang |
  619. |:---:|--------|:----:|------|
  620. | 01 | [Prompt Engineering: Techniques & Patterns](phases/11-llm-engineering/01-prompt-engineering/) | Build | Python, TypeScript |
  621. | 02 | [Few-Shot, CoT, Tree-of-Thought](phases/11-llm-engineering/02-few-shot-cot/) | Build | Python |
  622. | 03 | [Structured Outputs](phases/11-llm-engineering/03-structured-outputs/) | Build | Python, TypeScript |
  623. | 04 | [Embeddings & Vector Representations](phases/11-llm-engineering/04-embeddings/) | Build | Python, TypeScript |
  624. | 05 | [Context Engineering](phases/11-llm-engineering/05-context-engineering/) | Build | Python, TypeScript |
  625. | 06 | [RAG: Retrieval-Augmented Generation](phases/11-llm-engineering/06-rag/) | Build | Python, TypeScript |
  626. | 07 | [Advanced RAG: Chunking, Reranking](phases/11-llm-engineering/07-advanced-rag/) | Build | Python, TypeScript |
  627. | 08 | [Fine-Tuning with LoRA & QLoRA](phases/11-llm-engineering/08-fine-tuning-lora/) | Build | Python |
  628. | 09 | [Function Calling & Tool Use](phases/11-llm-engineering/09-function-calling/) | Build | Python, TypeScript |
  629. | 10 | [Evaluation & Testing](phases/11-llm-engineering/10-evaluation/) | Build | Python, TypeScript |
  630. | 11 | [Caching, Rate Limiting & Cost](phases/11-llm-engineering/11-caching-cost/) | Build | Python, TypeScript |
  631. | 12 | [Guardrails & Safety](phases/11-llm-engineering/12-guardrails/) | Build | Python, TypeScript |
  632. | 13 | [Building a Production LLM App](phases/11-llm-engineering/13-production-app/) | Build | Python, TypeScript |
  633. | 14 | [Model Context Protocol (MCP)](phases/11-llm-engineering/14-model-context-protocol/) | Build | Python, TypeScript, Go |
  634. | 15 | [Prompt Caching & Context Caching](phases/11-llm-engineering/15-prompt-caching/) | Build | Python, TypeScript |
  635. | 16 | [Agent State Machines — Graphs, Nodes, Checkpoints](phases/11-llm-engineering/16-langgraph-state-machines/) | Build | Python |
  636. | 17 | [Agent Framework Tradeoffs](phases/11-llm-engineering/17-agent-framework-tradeoffs/) | Learn | Python, TypeScript |
  637. </details>
  638. <details id="phase-12">
  639. <summary><b>Phase 12 — Multimodal AI</b> &nbsp;<code>25 lessons</code>&nbsp; <em>See, hear, read, and reason across modalities — from ViT patches to computer-use agents.</em></summary>
  640. <br/>
  641. | # | Lesson | Type | Lang |
  642. |:---:|--------|:----:|------|
  643. | 01 | [Vision Transformers and the Patch-Token Primitive](phases/12-multimodal-ai/01-vision-transformer-patch-tokens/) | Learn | Python, TypeScript |
  644. | 02 | [CLIP and Contrastive Vision-Language Pretraining](phases/12-multimodal-ai/02-clip-contrastive-pretraining/) | Build | Python, TypeScript |
  645. | 03 | [BLIP-2 Q-Former as Modality Bridge](phases/12-multimodal-ai/03-blip2-qformer-bridge/) | Build | Python, TypeScript |
  646. | 04 | [Flamingo and Gated Cross-Attention](phases/12-multimodal-ai/04-flamingo-gated-cross-attention/) | Learn | Python, TypeScript |
  647. | 05 | [LLaVA and Visual Instruction Tuning](phases/12-multimodal-ai/05-llava-visual-instruction-tuning/) | Build | Python, TypeScript |
  648. | 06 | [Any-Resolution Vision — Patch-n'-Pack and NaFlex](phases/12-multimodal-ai/06-any-resolution-patch-n-pack/) | Build | Python, TypeScript |
  649. | 07 | [Open-Weight VLM Recipes: What Actually Matters](phases/12-multimodal-ai/07-open-weight-vlm-recipes/) | Learn | Python, TypeScript |
  650. | 08 | [LLaVA-OneVision: Single, Multi, Video](phases/12-multimodal-ai/08-llava-onevision-single-multi-video/) | Build | Python, TypeScript |
  651. | 09 | [Qwen-VL Family and Dynamic-FPS Video](phases/12-multimodal-ai/09-qwen-vl-family-dynamic-fps/) | Learn | Python, TypeScript |
  652. | 10 | [InternVL3 Native Multimodal Pretraining](phases/12-multimodal-ai/10-internvl3-native-multimodal/) | Learn | Python, TypeScript |
  653. | 11 | [Chameleon Early-Fusion Token-Only](phases/12-multimodal-ai/11-chameleon-early-fusion-tokens/) | Build | Python, TypeScript |
  654. | 12 | [Emu3 Next-Token Prediction for Generation](phases/12-multimodal-ai/12-emu3-next-token-for-generation/) | Learn | Python, TypeScript |
  655. | 13 | [Transfusion Autoregressive + Diffusion](phases/12-multimodal-ai/13-transfusion-autoregressive-diffusion/) | Build | Python, TypeScript |
  656. | 14 | [Show-o Discrete-Diffusion Unified](phases/12-multimodal-ai/14-show-o-discrete-diffusion-unified/) | Learn | Python, TypeScript |
  657. | 15 | [Janus-Pro Decoupled Encoders](phases/12-multimodal-ai/15-janus-pro-decoupled-encoders/) | Build | Python, TypeScript |
  658. | 16 | [MIO Any-to-Any Streaming](phases/12-multimodal-ai/16-mio-any-to-any-streaming/) | Learn | Python, TypeScript |
  659. | 17 | [Video-Language Temporal Grounding](phases/12-multimodal-ai/17-video-language-temporal-grounding/) | Build | Python, TypeScript |
  660. | 18 | [Long-Video at Million-Token Context](phases/12-multimodal-ai/18-long-video-million-token/) | Build | Python, TypeScript |
  661. | 19 | [Audio-Language Models: Whisper to AF3](phases/12-multimodal-ai/19-audio-language-whisper-to-af3/) | Build | Python, TypeScript |
  662. | 20 | [Omni Models: Thinker-Talker Streaming](phases/12-multimodal-ai/20-omni-models-thinker-talker/) | Build | Python, TypeScript |
  663. | 21 | [Embodied VLAs: RT-2, OpenVLA, π0, GR00T](phases/12-multimodal-ai/21-embodied-vlas-openvla-pi0-groot/) | Learn | Python, TypeScript |
  664. | 22 | [Document and Diagram Understanding](phases/12-multimodal-ai/22-document-diagram-understanding/) | Build | Python, TypeScript |
  665. | 23 | [ColPali Vision-Native Document RAG](phases/12-multimodal-ai/23-colpali-vision-native-rag/) | Build | Python, TypeScript |
  666. | 24 | [Multimodal RAG and Cross-Modal Retrieval](phases/12-multimodal-ai/24-multimodal-rag-cross-modal/) | Build | Python, TypeScript |
  667. | 25 | [Multimodal Agents and Computer-Use (Capstone)](phases/12-multimodal-ai/25-multimodal-agents-computer-use/) | Build | Python, TypeScript |
  668. </details>
  669. <details id="phase-13">
  670. <summary><b>Phase 13 — Tools & Protocols</b> &nbsp;<code>45 lessons</code>&nbsp; <em>The interfaces between AI and the real world.</em></summary>
  671. <br/>
  672. | # | Lesson | Type | Lang |
  673. |:---:|--------|:----:|------|
  674. | 01 | [The Tool Interface](phases/13-tools-and-protocols/01-the-tool-interface/) | Learn | Python, TypeScript |
  675. | 02 | [Function Calling Deep Dive](phases/13-tools-and-protocols/02-function-calling-deep-dive/) | Build | Python, TypeScript |
  676. | 03 | [Parallel and Streaming Tool Calls](phases/13-tools-and-protocols/03-parallel-and-streaming-tool-calls/) | Build | Python, TypeScript, Go |
  677. | 04 | [Structured Output](phases/13-tools-and-protocols/04-structured-output/) | Build | Python, TypeScript |
  678. | 05 | [Tool Schema Design](phases/13-tools-and-protocols/05-tool-schema-design/) | Learn | Python, TypeScript |
  679. | 06 | [MCP Fundamentals: Stateless Requests and JSON-RPC](phases/13-tools-and-protocols/06-mcp-fundamentals/) | Learn | Python, TypeScript |
  680. | 07 | [Building an MCP Server: Stateless Python and TypeScript](phases/13-tools-and-protocols/07-building-an-mcp-server/) | Build | Python, TypeScript, Go |
  681. | 08 | [Building an MCP Client: Discovery, Routing, and Dual-Era Fallback](phases/13-tools-and-protocols/08-building-an-mcp-client/) | Build | Python, TypeScript |
  682. | 09 | [MCP Transports: stdio and Stateless Streamable HTTP](phases/13-tools-and-protocols/09-mcp-transports/) | Learn | Python, TypeScript, Go |
  683. | 10 | [MCP Resources and Prompts: Addressable Context for Stateless Servers](phases/13-tools-and-protocols/10-mcp-resources-and-prompts/) | Build | Python, TypeScript |
  684. | 11 | [MCP Model Input: Sampling Migration and Stateless MRTR](phases/13-tools-and-protocols/11-mcp-sampling/) | Build | Python, TypeScript |
  685. | 12 | [Explicit Scope and Stateless Elicitation](phases/13-tools-and-protocols/12-mcp-roots-and-elicitation/) | Build | Python, TypeScript |
  686. | 13 | [MCP Tasks Extension: Durable Work on a Stateless Core](phases/13-tools-and-protocols/13-mcp-async-tasks/) | Build | Python, TypeScript |
  687. | 14 | [MCP Apps on the Stateless Protocol](phases/13-tools-and-protocols/14-mcp-apps/) | Build | Python, TypeScript |
  688. | 15 | [MCP Security: Poisoned Metadata, Routing, and MRTR State](phases/13-tools-and-protocols/15-mcp-security-tool-poisoning/) | Learn | Python, TypeScript |
  689. | 16 | [MCP Authorization: CIMD, Issuer Binding, PKCE, and Step-Up](phases/13-tools-and-protocols/16-mcp-security-oauth-2-1/) | Build | Python, TypeScript, Go |
  690. | 17 | [Stateless MCP Gateways and Registry Admission](phases/13-tools-and-protocols/17-mcp-gateways-and-registries/) | Learn | Python, TypeScript, Go |
  691. | 18 | [MCP Auth in Production: Issuer-Bound Enrollment and Tokens](phases/13-tools-and-protocols/18-mcp-auth-production/) | Build | Python, TypeScript, Go |
  692. | 19 | [A2A Protocol](phases/13-tools-and-protocols/19-a2a-protocol/) | Build | Python, TypeScript, Go |
  693. | 20 | [OpenTelemetry GenAI](phases/13-tools-and-protocols/20-opentelemetry-genai/) | Build | Python, TypeScript |
  694. | 21 | [LLM Routing Layer](phases/13-tools-and-protocols/21-llm-routing-layer/) | Learn | Python, TypeScript, Go |
  695. | 22 | [Agent Skills: Portable Contract and Runtime Boundary](phases/13-tools-and-protocols/22-skills-and-agent-sdks/) | Build | Python, TypeScript |
  696. | 23 | [Capstone: Stateless Tool Ecosystem](phases/13-tools-and-protocols/23-capstone-tool-ecosystem/) | Build | Python, TypeScript |
  697. | 24 | [Skill Discovery and Progressive Disclosure](phases/13-tools-and-protocols/24-skill-discovery-and-progressive-disclosure/) | Build | Python, TypeScript |
  698. | 25 | [Skill Invocation and Routing](phases/13-tools-and-protocols/25-skill-invocation-and-routing/) | Build | Python, TypeScript |
  699. | 26 | [Skill Permissions, Sandboxes, and Trust](phases/13-tools-and-protocols/26-skill-permissions-sandboxes-and-trust/) | Build | Python, TypeScript |
  700. | 27 | [Skill Evals, Packaging, and Portability](phases/13-tools-and-protocols/27-skill-evals-packaging-and-portability/) | Build | Python, TypeScript |
  701. | 28 | [MCP Tool Contracts and Content](phases/13-tools-and-protocols/28-mcp-tool-contracts-and-content/) | Build | Python, TypeScript |
  702. | 29 | [MCP Reliability, Cancellation, and Flow Control](phases/13-tools-and-protocols/29-mcp-reliability-cancellation-and-flow-control/) | Build | Python, TypeScript |
  703. | 30 | [MCP Registry Supply Chain: Admission, Drift, and Rollback](phases/13-tools-and-protocols/30-mcp-registry-supply-chain-and-drift/) | Build | Python, TypeScript |
  704. | 31 | [MCP Conformance Engineering: Versioning, Evidence, and Operations](phases/13-tools-and-protocols/31-mcp-conformance-versioning-and-operations/) | Build | Python, TypeScript |
  705. | 32 | [Program Anatomy: An ELF Reader From Scratch](phases/13-tools-and-protocols/32-program-anatomy-elf-from-scratch/) | Build | Python, TypeScript, Rust |
  706. | 33 | [Disassembly and Control Flow: From Bytes to a Graph You Can Check](phases/13-tools-and-protocols/33-disassembly-and-control-flow/) | Build | Python, TypeScript, Rust |
  707. | 34 | [Lifting to Pseudocode: IR, SSA, and Structuring From Scratch](phases/13-tools-and-protocols/34-lifting-to-pseudocode/) | Build | Python, TypeScript |
  708. | 35 | [Evidence, Confidence, and Unknowns in Binary Analysis](phases/13-tools-and-protocols/35-evidence-confidence-and-unknowns/) | Build | Python, TypeScript |
  709. | 36 | [Rebuild and Prove Equal: Differential Tests, Byte Comparison, and the Evidence Loop](phases/13-tools-and-protocols/36-rebuild-and-prove-equal/) | Build | Python, TypeScript |
  710. | 37 | [The Reverse Engineering Landscape: History, Tool Families, and How a Decompiler Is Built](phases/13-tools-and-protocols/37-the-reverse-engineering-landscape/) | Build | Python, TypeScript |
  711. | 38 | [Calling Conventions and Stack Frames: System V, Windows x64, and AArch64](phases/13-tools-and-protocols/38-calling-conventions-and-stack-frames/) | Build | Python, TypeScript |
  712. | 39 | [Compiler Idioms: What Optimization Does to Your Code and How to Read It Back](phases/13-tools-and-protocols/39-compiler-idioms-and-optimization/) | Build | Python, TypeScript |
  713. | 40 | [Linking, Loading, and Stripped Binaries: PLT, GOT, Imports, Fixups, and Function Identification](phases/13-tools-and-protocols/40-linking-loading-and-stripped-binaries/) | Build | Python, TypeScript, Rust |
  714. | 41 | [Recovering Data and Types: Structs, Arrays, C++ Objects, and DWARF](phases/13-tools-and-protocols/41-recovering-data-and-types/) | Build | Python, TypeScript |
  715. | 42 | [Dynamic Analysis of Your Own Program: Debuggers, Traces, and Confirming Static Claims](phases/13-tools-and-protocols/42-dynamic-analysis-of-your-own-program/) | Build | Python, TypeScript |
  716. | 43 | [Beyond Native Code: Python, JVM, .NET, and WebAssembly Bytecode](phases/13-tools-and-protocols/43-bytecode-and-managed-formats/) | Build | Python, TypeScript |
  717. | 44 | [Binary Diffing and Symbolic Execution: Matching Functions and Solving for Inputs](phases/13-tools-and-protocols/44-binary-diffing-and-symbolic-execution/) | Build | Python, TypeScript |
  718. | 45 | [AI-Assisted Reverse Engineering: LLM Decompilers, Analysis Agents, and Ground Truth](phases/13-tools-and-protocols/45-ai-assisted-reverse-engineering/) | Build | Python, TypeScript |
  719. Lessons 06-18 and 28-31 form the focused
  720. [Model Context Protocol (MCP) path](learning-paths/model-context-protocol.json). Its manifest order
  721. is 06, 07, 08, 09, 10, 11, 12, 13, 14, 15, 16, 18, 17, 28, 29, 30, 31. Start
  722. it with the host-specific `learn-mcp` invocation above. Lesson 23
  723. is its only optional capstone and also requires Lessons 19 and 20.
  724. Lessons 22 and 24-27 form the focused
  725. [Agent Skills learning path](learning-paths/agent-skills.json), from package
  726. contract through real-host release gates. Start it with the host-specific
  727. `learn-agent-skills` invocation shown above; do not follow numeric next
  728. navigation from 22 to 23.
  729. </details>
  730. <details id="phase-14">
  731. <summary><b>Phase 14 — Agent Engineering</b> &nbsp;<code>54 lessons</code>&nbsp; <em>Build agents from first principles, use coding agents reliably, and shape the work before implementation.</em></summary>
  732. <br/>
  733. | # | Lesson | Type | Lang |
  734. |:---:|--------|:----:|------|
  735. | 01 | [The Agent Loop](phases/14-agent-engineering/01-the-agent-loop/) | Build | Python, TypeScript |
  736. | 02 | [ReWOO and Plan-and-Execute](phases/14-agent-engineering/02-rewoo-plan-and-execute/) | Build | Python, TypeScript |
  737. | 03 | [Reflexion and Verbal Reinforcement Learning](phases/14-agent-engineering/03-reflexion-verbal-rl/) | Build | Python, TypeScript |
  738. | 04 | [Tree of Thoughts and LATS](phases/14-agent-engineering/04-tree-of-thoughts-lats/) | Build | Python, TypeScript |
  739. | 05 | [Self-Refine and CRITIC](phases/14-agent-engineering/05-self-refine-and-critic/) | Build | Python, TypeScript |
  740. | 06 | [Tool Use and Function Calling](phases/14-agent-engineering/06-tool-use-and-function-calling/) | Build | Python, TypeScript |
  741. | 07 | [Agent Memory — Virtual Context and Memory Paging](phases/14-agent-engineering/07-memory-virtual-context-memgpt/) | Build | Python, TypeScript |
  742. | 08 | [Memory Blocks and Sleep-Time Compute](phases/14-agent-engineering/08-memory-blocks-sleep-time-compute/) | Build | Python, TypeScript |
  743. | 09 | [Hybrid Memory — Vector + Graph + KV](phases/14-agent-engineering/09-hybrid-memory-mem0/) | Build | Python, TypeScript |
  744. | 10 | [Skill Libraries and Lifelong Learning (Voyager)](phases/14-agent-engineering/10-skill-libraries-voyager/) | Build | Python, TypeScript |
  745. | 11 | [Planning with HTN and Evolutionary Search](phases/14-agent-engineering/11-planning-htn-and-evolutionary/) | Build | Python, TypeScript |
  746. | 12 | [Anthropic's Workflow Patterns](phases/14-agent-engineering/12-anthropic-workflow-patterns/) | Build | Python, TypeScript |
  747. | 13 | [Stateful Graph Orchestration — Durable Execution and Checkpoints](phases/14-agent-engineering/13-langgraph-stateful-graphs/) | Build | Python, TypeScript |
  748. | 14 | [The Actor Model for Agents](phases/14-agent-engineering/14-autogen-actor-model/) | Build | Python, TypeScript |
  749. | 15 | [Role-Based Agent Teams — Roles, Tasks, Processes](phases/14-agent-engineering/15-crewai-role-based-crews/) | Build | Python |
  750. | 16 | [OpenAI Agents SDK — Handoffs, Guardrails, Tracing](phases/14-agent-engineering/16-openai-agents-sdk/) | Build | Python, TypeScript |
  751. | 17 | [The Harness as a Library — Subagents and Session Store](phases/14-agent-engineering/17-claude-agent-sdk/) | Build | Python, TypeScript |
  752. | 18 | [Production Agent Runtimes](phases/14-agent-engineering/18-agno-and-mastra-runtimes/) | Learn | Python, TypeScript |
  753. | 19 | [Benchmarks — SWE-bench, GAIA, AgentBench](phases/14-agent-engineering/19-benchmarks-swebench-gaia/) | Learn | Python, TypeScript |
  754. | 20 | [Benchmarks — WebArena and OSWorld](phases/14-agent-engineering/20-benchmarks-webarena-osworld/) | Learn | Python, TypeScript |
  755. | 21 | [Computer Use — Claude, OpenAI CUA, Gemini](phases/14-agent-engineering/21-computer-use-agents/) | Build | Python, TypeScript |
  756. | 22 | [Voice Agents — Pipecat and LiveKit](phases/14-agent-engineering/22-voice-agents-pipecat-livekit/) | Build | Python, TypeScript |
  757. | 23 | [OpenTelemetry GenAI Semantic Conventions](phases/14-agent-engineering/23-otel-genai-conventions/) | Build | Python, TypeScript, Go |
  758. | 24 | [Agent Observability — Langfuse, Phoenix, Opik](phases/14-agent-engineering/24-agent-observability-platforms/) | Learn | Python, TypeScript |
  759. | 25 | [Multi-Agent Debate and Collaboration](phases/14-agent-engineering/25-multi-agent-debate/) | Build | Python, TypeScript |
  760. | 26 | [Failure Modes — Why Agents Break](phases/14-agent-engineering/26-failure-modes-agentic/) | Build | Python, TypeScript |
  761. | 27 | [Prompt Injection and the PVE Defense](phases/14-agent-engineering/27-prompt-injection-defense/) | Build | Python, TypeScript |
  762. | 28 | [Orchestration Patterns — Supervisor, Swarm, Hierarchical](phases/14-agent-engineering/28-orchestration-patterns/) | Build | Python, TypeScript |
  763. | 29 | [Production Runtimes — Queue, Event, Cron](phases/14-agent-engineering/29-production-runtimes/) | Learn | Python, TypeScript, Go |
  764. | 30 | [Eval-Driven Agent Development](phases/14-agent-engineering/30-eval-driven-agent-development/) | Build | Python, TypeScript |
  765. | 31 | [Agent Workbench: Why Capable Models Still Fail](phases/14-agent-engineering/31-agent-workbench-why-models-fail/) | Learn | Python, TypeScript |
  766. | 32 | [The Minimal Agent Workbench](phases/14-agent-engineering/32-minimal-agent-workbench/) | Build | Python, TypeScript |
  767. | 33 | [Agent Instructions as Executable Constraints](phases/14-agent-engineering/33-instructions-as-executable-constraints/) | Build | Python, TypeScript |
  768. | 34 | [Repo Memory and Durable State](phases/14-agent-engineering/34-repo-memory-and-state/) | Build | Python, TypeScript |
  769. | 35 | [Initialization Scripts for Agents](phases/14-agent-engineering/35-initialization-scripts/) | Build | Python, TypeScript |
  770. | 36 | [Scope Contracts and Task Boundaries](phases/14-agent-engineering/36-scope-contracts/) | Build | Python, TypeScript |
  771. | 37 | [Runtime Feedback Loops](phases/14-agent-engineering/37-runtime-feedback-loops/) | Build | Python, TypeScript |
  772. | 38 | [Verification Gates](phases/14-agent-engineering/38-verification-gates/) | Build | Python, TypeScript |
  773. | 39 | [Reviewer Agent: Separate Builder from Marker](phases/14-agent-engineering/39-reviewer-agent/) | Build | Python, TypeScript |
  774. | 40 | [Multi-Session Handoff](phases/14-agent-engineering/40-multi-session-handoff/) | Build | Python, TypeScript |
  775. | 41 | [The Workbench on a Real Repo](phases/14-agent-engineering/41-workbench-for-real-repos/) | Build | Python, TypeScript |
  776. | 42 | [Capstone: Ship a Reusable Agent Workbench Pack](phases/14-agent-engineering/42-agent-workbench-capstone/) | Build | Python, TypeScript |
  777. | 43 | [Frame the Task Before the Agent Writes Code](phases/14-agent-engineering/43-frame-the-task-before-code/) | Build | Python, TypeScript |
  778. | 44 | [Build an Evidence-Backed Execution Plan](phases/14-agent-engineering/44-plan-from-evidence/) | Build | Python, TypeScript |
  779. | 45 | [Delegate Agent Work with Isolation and Merge Contracts](phases/14-agent-engineering/45-delegate-with-isolation/) | Build | Python, TypeScript |
  780. | 46 | [Turn Every Agent Correction into a System Improvement](phases/14-agent-engineering/46-turn-feedback-into-system/) | Build | Python, TypeScript |
  781. | 47 | [Define the Outcome Before You Choose the Output](phases/14-agent-engineering/47-outcomes-before-output/) | Build | Python, TypeScript |
  782. | 48 | [Discover the Workflow People Actually Perform](phases/14-agent-engineering/48-discover-the-real-workflow/) | Build | Python, TypeScript |
  783. | 49 | [Map Assumptions and Resolve the Riskiest One First](phases/14-agent-engineering/49-map-assumptions-and-risk/) | Build | Python, TypeScript |
  784. | 50 | [Choose the Smallest Slice That Can Change the Decision](phases/14-agent-engineering/50-choose-the-smallest-testable-slice/) | Build | Python, TypeScript |
  785. | 51 | [Write Specifications That Preserve Judgment](phases/14-agent-engineering/51-write-specifications-that-preserve-judgment/) | Build | Python, TypeScript |
  786. | 52 | [Design Success Metrics Before the Result Exists](phases/14-agent-engineering/52-design-success-metrics/) | Build | Python, TypeScript |
  787. | 53 | [Choose Prototype, Pilot, or Production Deliberately](phases/14-agent-engineering/53-prototype-pilot-or-production/) | Build | Python, TypeScript |
  788. | 54 | [Build a Feedback Ratchet with Ownership and Retirement](phases/14-agent-engineering/54-build-the-feedback-ratchet/) | Build | Python, TypeScript |
  789. Each Phase 14 workbench lesson (31-42) ships a `mission.md` briefing the agent before it opens the full lesson docs.
  790. Lessons 31-46 form the [Agent-Assisted Engineering path](learning-paths/using-coding-agents.json).
  791. Its manifest order combines the workbench foundation with task framing, planning,
  792. delegation, and durable feedback. Lessons 47-54 form the
  793. [Product Judgment and Delivery path](learning-paths/shaping-the-build.json), from outcome framing
  794. through evidence, risk, scope, measurement, staged release, and feedback ownership.
  795. </details>
  796. <details id="phase-15">
  797. <summary><b>Phase 15 — Autonomous Systems</b> &nbsp;<code>22 lessons</code>&nbsp; <em>Long-horizon agents, self-improvement, and the 2026 safety stack.</em></summary>
  798. <br/>
  799. | # | Lesson | Type | Lang |
  800. |:---:|--------|:----:|------|
  801. | 01 | [From Chatbots to Long-Horizon Agents (METR)](phases/15-autonomous-systems/01-long-horizon-agents/) | Learn | Python, TypeScript |
  802. | 02 | [STaR, V-STaR, Quiet-STaR: Self-Taught Reasoning](phases/15-autonomous-systems/02-star-family-reasoning/) | Learn | Python, TypeScript |
  803. | 03 | [AlphaEvolve: Evolutionary Coding Agents](phases/15-autonomous-systems/03-alphaevolve-evolutionary-coding/) | Learn | Python, TypeScript |
  804. | 04 | [Darwin Gödel Machine: Self-Modifying Agents](phases/15-autonomous-systems/04-darwin-godel-machine/) | Learn | Python, TypeScript |
  805. | 05 | [AI Scientist v2: Workshop-Level Research](phases/15-autonomous-systems/05-ai-scientist-v2/) | Learn | Python, TypeScript |
  806. | 06 | [Automated Alignment Research (Anthropic AAR)](phases/15-autonomous-systems/06-automated-alignment-research/) | Learn | Python, TypeScript |
  807. | 07 | [Recursive Self-Improvement: Capability vs Alignment](phases/15-autonomous-systems/07-recursive-self-improvement/) | Learn | Python, TypeScript |
  808. | 08 | [Bounded Self-Improvement Designs](phases/15-autonomous-systems/08-bounded-self-improvement/) | Learn | Python, TypeScript |
  809. | 09 | [Autonomous Coding Agent Landscape (SWE-bench, CodeAct)](phases/15-autonomous-systems/09-coding-agent-landscape/) | Learn | Python, TypeScript |
  810. | 10 | [Permission Modes for Autonomous Agents](phases/15-autonomous-systems/10-claude-code-permission-modes/) | Learn | Python, TypeScript |
  811. | 11 | [Browser Agents and Indirect Prompt Injection](phases/15-autonomous-systems/11-browser-agents/) | Learn | Python, TypeScript |
  812. | 12 | [Durable Execution for Long-Running Agents](phases/15-autonomous-systems/12-durable-execution/) | Learn | Python, TypeScript, Go |
  813. | 13 | [Action Budgets, Iteration Caps, Cost Governors](phases/15-autonomous-systems/13-cost-governors/) | Learn | Python, TypeScript |
  814. | 14 | [Kill Switches, Circuit Breakers, Canary Tokens](phases/15-autonomous-systems/14-kill-switches-canaries/) | Learn | Python, TypeScript |
  815. | 15 | [HITL: Propose-Then-Commit](phases/15-autonomous-systems/15-propose-then-commit/) | Learn | Python, TypeScript |
  816. | 16 | [Checkpoints and Rollback](phases/15-autonomous-systems/16-checkpoints-rollback/) | Learn | Python, TypeScript |
  817. | 17 | [Constitutional AI and Rule Overrides](phases/15-autonomous-systems/17-constitutional-ai/) | Learn | Python, TypeScript |
  818. | 18 | [Llama Guard and Input/Output Classification](phases/15-autonomous-systems/18-llama-guard/) | Learn | Python, TypeScript |
  819. | 19 | [Anthropic Responsible Scaling Policy v3.0](phases/15-autonomous-systems/19-anthropic-rsp/) | Learn | Python, TypeScript |
  820. | 20 | [OpenAI Preparedness Framework and DeepMind FSF](phases/15-autonomous-systems/20-openai-preparedness-deepmind-fsf/) | Learn | Python, TypeScript |
  821. | 21 | [METR Time Horizons and External Evaluation](phases/15-autonomous-systems/21-metr-external-evaluation/) | Learn | Python, TypeScript, Julia |
  822. | 22 | [CAIS, CAISI, and Societal-Scale Risk](phases/15-autonomous-systems/22-cais-caisi-societal-risk/) | Learn | Python, TypeScript |
  823. </details>
  824. <details id="phase-16">
  825. <summary><b>Phase 16 — Multi-Agent & Swarms</b> &nbsp;<code>25 lessons</code>&nbsp; <em>Coordination, emergence, and collective intelligence.</em></summary>
  826. <br/>
  827. | # | Lesson | Type | Lang |
  828. |:---:|--------|:----:|------|
  829. | 01 | [Why Multi-Agent](phases/16-multi-agent-and-swarms/01-why-multi-agent/) | Learn | TypeScript |
  830. | 02 | [FIPA-ACL Heritage and Speech Acts](phases/16-multi-agent-and-swarms/02-fipa-acl-heritage/) | Learn | Python, TypeScript |
  831. | 03 | [Communication Protocols](phases/16-multi-agent-and-swarms/03-communication-protocols/) | Build | TypeScript |
  832. | 04 | [The Multi-Agent Primitive Model](phases/16-multi-agent-and-swarms/04-primitive-model/) | Learn | Python, TypeScript |
  833. | 05 | [Supervisor / Orchestrator-Worker Pattern](phases/16-multi-agent-and-swarms/05-supervisor-orchestrator-pattern/) | Build | Python, TypeScript |
  834. | 06 | [Hierarchical Architecture and Decomposition Drift](phases/16-multi-agent-and-swarms/06-hierarchical-architecture/) | Learn | Python, TypeScript |
  835. | 07 | [Society of Mind and Multi-Agent Debate](phases/16-multi-agent-and-swarms/07-society-of-mind-debate/) | Build | Python, TypeScript |
  836. | 08 | [Role Specialization — Planner / Critic / Executor / Verifier](phases/16-multi-agent-and-swarms/08-role-specialization/) | Build | Python, TypeScript |
  837. | 09 | [Parallel Swarm and Networked Architectures](phases/16-multi-agent-and-swarms/09-parallel-swarm-networks/) | Build | Python, TypeScript, Go |
  838. | 10 | [Group Chat and Speaker Selection](phases/16-multi-agent-and-swarms/10-group-chat-speaker-selection/) | Build | Python, TypeScript |
  839. | 11 | [Handoffs and Routines (Stateless Orchestration)](phases/16-multi-agent-and-swarms/11-handoffs-and-routines/) | Build | Python, TypeScript |
  840. | 12 | [A2A — The Agent-to-Agent Protocol](phases/16-multi-agent-and-swarms/12-a2a-protocol/) | Build | Python, TypeScript, Go |
  841. | 13 | [Shared Memory and Blackboard Patterns](phases/16-multi-agent-and-swarms/13-shared-memory-blackboard/) | Build | Python, TypeScript |
  842. | 14 | [Consensus and Byzantine Fault Tolerance](phases/16-multi-agent-and-swarms/14-consensus-and-bft/) | Build | Python, TypeScript |
  843. | 15 | [Voting, Self-Consistency, and Debate Topology](phases/16-multi-agent-and-swarms/15-voting-debate-topology/) | Build | Python, TypeScript |
  844. | 16 | [Negotiation and Bargaining](phases/16-multi-agent-and-swarms/16-negotiation-bargaining/) | Build | Python, TypeScript |
  845. | 17 | [Generative Agents and Emergent Simulation](phases/16-multi-agent-and-swarms/17-generative-agents-simulation/) | Build | Python, TypeScript |
  846. | 18 | [Theory of Mind and Emergent Coordination](phases/16-multi-agent-and-swarms/18-theory-of-mind-coordination/) | Build | Python, TypeScript |
  847. | 19 | [Swarm Optimization (PSO, ACO)](phases/16-multi-agent-and-swarms/19-swarm-optimization-pso-aco/) | Build | Python, TypeScript, Julia |
  848. | 20 | [MARL — MADDPG, QMIX, MAPPO](phases/16-multi-agent-and-swarms/20-marl-maddpg-qmix-mappo/) | Learn | Python, TypeScript |
  849. | 21 | [Agent Economies, Token Incentives, Reputation](phases/16-multi-agent-and-swarms/21-agent-economies/) | Learn | Python, TypeScript |
  850. | 22 | [Production Scaling — Queues, Checkpoints, Durability](phases/16-multi-agent-and-swarms/22-production-scaling-queues-checkpoints/) | Build | Python, TypeScript |
  851. | 23 | [Failure Modes — MAST, Groupthink, Monoculture](phases/16-multi-agent-and-swarms/23-failure-modes-mast-groupthink/) | Learn | Python, TypeScript |
  852. | 24 | [Evaluation and Coordination Benchmarks](phases/16-multi-agent-and-swarms/24-evaluation-coordination-benchmarks/) | Learn | Python, TypeScript |
  853. | 25 | [Case Studies and 2026 State of the Art](phases/16-multi-agent-and-swarms/25-case-studies-2026-sota/) | Learn | Python, TypeScript |
  854. </details>
  855. <details id="phase-17">
  856. <summary><b>Phase 17 — Infrastructure & Production</b> &nbsp;<code>29 lessons</code>&nbsp; <em>Ship AI to the real world.</em></summary>
  857. <br/>
  858. | # | Lesson | Type | Lang |
  859. |:---:|--------|:----:|------|
  860. | 01 | [Managed LLM Platforms — Bedrock, Azure OpenAI, Vertex AI](phases/17-infrastructure-and-production/01-managed-llm-platforms/) | Learn | Python, TypeScript |
  861. | 02 | [Inference Platform Economics — Fireworks, Together, Baseten, Modal](phases/17-infrastructure-and-production/02-inference-platform-economics/) | Learn | Python, TypeScript |
  862. | 03 | [GPU Autoscaling on Kubernetes — Karpenter, KAI Scheduler](phases/17-infrastructure-and-production/03-gpu-autoscaling-kubernetes/) | Learn | Python, TypeScript |
  863. | 04 | [Serving Engine Internals — PagedAttention, Continuous Batching, Chunked Prefill](phases/17-infrastructure-and-production/04-vllm-serving-internals/) | Learn | Python, TypeScript, Rust |
  864. | 05 | [EAGLE-3 Speculative Decoding in Production](phases/17-infrastructure-and-production/05-eagle3-speculative-decoding/) | Learn | Python, TypeScript |
  865. | 06 | [Prefix-Cache Serving — RadixAttention and KV Reuse](phases/17-infrastructure-and-production/06-sglang-radixattention/) | Learn | Python, TypeScript, Rust |
  866. | 07 | [Hardware-Specialized Inference Compilation — FP8 and NVFP4 on Blackwell](phases/17-infrastructure-and-production/07-tensorrt-llm-blackwell/) | Learn | Python, TypeScript |
  867. | 08 | [Inference Metrics — TTFT, TPOT, ITL, Goodput, P99](phases/17-infrastructure-and-production/08-inference-metrics-goodput/) | Learn | Python, TypeScript |
  868. | 09 | [Production Quantization — AWQ, GPTQ, GGUF, FP8, NVFP4](phases/17-infrastructure-and-production/09-production-quantization/) | Learn | Python, TypeScript |
  869. | 10 | [Cold Start Mitigation for Serverless LLMs](phases/17-infrastructure-and-production/10-cold-start-mitigation/) | Learn | Python, TypeScript |
  870. | 11 | [Multi-Region LLM Serving and KV Cache Locality](phases/17-infrastructure-and-production/11-multi-region-kv-locality/) | Learn | Python, TypeScript |
  871. | 12 | [Edge Inference — ANE, Hexagon, WebGPU, Jetson](phases/17-infrastructure-and-production/12-edge-inference/) | Learn | Python, TypeScript |
  872. | 13 | [LLM Observability Stack Selection](phases/17-infrastructure-and-production/13-llm-observability/) | Learn | Python, TypeScript |
  873. | 14 | [Prompt Caching and Semantic Caching Economics](phases/17-infrastructure-and-production/14-prompt-semantic-caching/) | Learn | Python, TypeScript |
  874. | 15 | [Batch APIs — the 50% Discount as Industry Standard](phases/17-infrastructure-and-production/15-batch-apis/) | Learn | Python, TypeScript |
  875. | 16 | [Model Routing as a Cost-Reduction Primitive](phases/17-infrastructure-and-production/16-model-routing/) | Learn | Python, TypeScript |
  876. | 17 | [Disaggregated Prefill/Decode — NVIDIA Dynamo and llm-d](phases/17-infrastructure-and-production/17-disaggregated-prefill-decode/) | Learn | Python, TypeScript |
  877. | 18 | [Production Serving Stack — KV Offloading and Cache-Aware Routing](phases/17-infrastructure-and-production/18-vllm-production-stack-lmcache/) | Learn | Python, TypeScript |
  878. | 19 | [AI Gateways — LiteLLM, Portkey, Kong, Bifrost](phases/17-infrastructure-and-production/19-ai-gateways/) | Learn | Python, TypeScript |
  879. | 20 | [Shadow, Canary, and Progressive Deployment](phases/17-infrastructure-and-production/20-shadow-canary-progressive/) | Learn | Python, TypeScript |
  880. | 21 | [A/B Testing LLM Features — GrowthBook and Statsig](phases/17-infrastructure-and-production/21-ab-testing-llm-features/) | Learn | Python, TypeScript, Julia |
  881. | 22 | [Load Testing LLM APIs — k6, LLMPerf, GenAI-Perf](phases/17-infrastructure-and-production/22-load-testing-llm-apis/) | Build | Python, TypeScript |
  882. | 23 | [SRE for AI — Multi-Agent Incident Response](phases/17-infrastructure-and-production/23-sre-for-ai/) | Learn | Python, TypeScript |
  883. | 24 | [Chaos Engineering for LLM Production](phases/17-infrastructure-and-production/24-chaos-engineering-llm/) | Learn | Python, TypeScript |
  884. | 25 | [Security — Secrets, PII Scrubbing, Audit Logs](phases/17-infrastructure-and-production/25-security-secrets-audit/) | Learn | Python, TypeScript |
  885. | 26 | [Compliance — SOC 2, HIPAA, GDPR, EU AI Act, ISO 42001](phases/17-infrastructure-and-production/26-compliance-frameworks/) | Learn | Python, TypeScript |
  886. | 27 | [FinOps for LLMs — Unit Economics and Multi-Tenant Attribution](phases/17-infrastructure-and-production/27-finops-llms/) | Learn | Python, TypeScript |
  887. | 28 | [Self-Hosted Serving Selection — Matching Engine to Hardware and Scale](phases/17-infrastructure-and-production/28-self-hosted-serving-selection/) | Learn | Python, TypeScript |
  888. | 29 | [Model Demo Interfaces — From a Function to a Shareable App](phases/17-infrastructure-and-production/29-model-demo-interfaces/) | Build | Python, TypeScript |
  889. </details>
  890. <details id="phase-18">
  891. <summary><b>Phase 18 — Ethics, Safety & Alignment</b> &nbsp;<code>30 lessons</code>&nbsp; <em>Build AI that helps humanity. Not optional.</em></summary>
  892. <br/>
  893. | # | Lesson | Type | Lang |
  894. |:---:|--------|:----:|------|
  895. | 01 | [Instruction-Following as Alignment Signal](phases/18-ethics-safety-alignment/01-instruction-following-alignment-signal/) | Learn | Python, TypeScript |
  896. | 02 | [Reward Hacking & Goodhart's Law](phases/18-ethics-safety-alignment/02-reward-hacking-goodhart/) | Learn | Python, TypeScript |
  897. | 03 | [Direct Preference Optimization Family](phases/18-ethics-safety-alignment/03-direct-preference-optimization-family/) | Learn | Python, TypeScript |
  898. | 04 | [Sycophancy as RLHF Amplification](phases/18-ethics-safety-alignment/04-sycophancy-rlhf-amplification/) | Learn | Python, TypeScript |
  899. | 05 | [Constitutional AI & RLAIF](phases/18-ethics-safety-alignment/05-constitutional-ai-rlaif/) | Learn | Python, TypeScript |
  900. | 06 | [Mesa-Optimization & Deceptive Alignment](phases/18-ethics-safety-alignment/06-mesa-optimization-deceptive-alignment/) | Learn | Python, TypeScript |
  901. | 07 | [Sleeper Agents — Persistent Deception](phases/18-ethics-safety-alignment/07-sleeper-agents-persistent-deception/) | Learn | Python, TypeScript |
  902. | 08 | [In-Context Scheming in Frontier Models](phases/18-ethics-safety-alignment/08-in-context-scheming-frontier-models/) | Learn | Python, TypeScript |
  903. | 09 | [Alignment Faking](phases/18-ethics-safety-alignment/09-alignment-faking/) | Learn | Python, TypeScript |
  904. | 10 | [AI Control — Safety Despite Subversion](phases/18-ethics-safety-alignment/10-ai-control-subversion/) | Learn | Python, TypeScript |
  905. | 11 | [Scalable Oversight & Weak-to-Strong](phases/18-ethics-safety-alignment/11-scalable-oversight-weak-to-strong/) | Learn | Python, TypeScript |
  906. | 12 | [Red-Teaming: PAIR & Automated Attacks](phases/18-ethics-safety-alignment/12-red-teaming-pair-automated-attacks/) | Build | Python, TypeScript |
  907. | 13 | [Many-Shot Jailbreaking](phases/18-ethics-safety-alignment/13-many-shot-jailbreaking/) | Learn | Python, TypeScript |
  908. | 14 | [ASCII Art & Visual Jailbreaks](phases/18-ethics-safety-alignment/14-ascii-art-visual-jailbreaks/) | Build | Python, TypeScript |
  909. | 15 | [Indirect Prompt Injection](phases/18-ethics-safety-alignment/15-indirect-prompt-injection/) | Build | Python, TypeScript |
  910. | 16 | [Red-Team Tooling: Garak, Llama Guard, PyRIT](phases/18-ethics-safety-alignment/16-red-team-tooling-garak-llamaguard-pyrit/) | Build | Python, TypeScript |
  911. | 17 | [WMDP & Dual-Use Capability Evaluation](phases/18-ethics-safety-alignment/17-wmdp-dual-use-evaluation/) | Learn | Python, TypeScript |
  912. | 18 | [Frontier Safety Frameworks — RSP, PF, FSF](phases/18-ethics-safety-alignment/18-frontier-safety-frameworks-rsp-pf-fsf/) | Learn | Python, TypeScript |
  913. | 19 | [Model Welfare Research](phases/18-ethics-safety-alignment/19-model-welfare-research/) | Learn | Python, TypeScript |
  914. | 20 | [Bias & Representational Harm](phases/18-ethics-safety-alignment/20-bias-representational-harm/) | Build | Python, TypeScript |
  915. | 21 | [Fairness Criteria: Group, Individual, Counterfactual](phases/18-ethics-safety-alignment/21-fairness-criteria-group-individual-counterfactual/) | Learn | Python, TypeScript |
  916. | 22 | [Differential Privacy for LLMs](phases/18-ethics-safety-alignment/22-differential-privacy-for-llms/) | Build | Python, TypeScript, Julia |
  917. | 23 | [Watermarking: SynthID, Stable Signature, C2PA](phases/18-ethics-safety-alignment/23-watermarking-synthid-stable-signature-c2pa/) | Build | Python, TypeScript |
  918. | 24 | [Regulatory Frameworks: EU, US, UK, Korea](phases/18-ethics-safety-alignment/24-regulatory-frameworks-eu-us-uk-korea/) | Learn | Python, TypeScript |
  919. | 25 | [EchoLeak & CVEs for AI](phases/18-ethics-safety-alignment/25-echoleak-cves-for-ai/) | Learn | Python, TypeScript |
  920. | 26 | [Model, System & Dataset Cards](phases/18-ethics-safety-alignment/26-model-system-dataset-cards/) | Build | Python, TypeScript |
  921. | 27 | [Data Provenance & Training-Data Governance](phases/18-ethics-safety-alignment/27-data-provenance-training-governance/) | Learn | Python, TypeScript |
  922. | 28 | [Alignment Research Ecosystem: MATS, Redwood, Apollo, METR](phases/18-ethics-safety-alignment/28-alignment-research-ecosystem/) | Learn | Python, TypeScript |
  923. | 29 | [Moderation Systems: OpenAI, Perspective, Llama Guard](phases/18-ethics-safety-alignment/29-moderation-systems-openai-perspective-llamaguard/) | Build | Python, TypeScript |
  924. | 30 | [Dual-Use Risk: Cyber, Bio, Chem, Nuclear](phases/18-ethics-safety-alignment/30-dual-use-risk-cyber-bio-chem-nuclear/) | Learn | Python, TypeScript |
  925. </details>
  926. <details id="phase-19">
  927. <summary><b>Phase 19 — Capstone Projects</b> &nbsp;<code>85 lessons</code>&nbsp; <em>17 end-to-end products + 9 deep-build tracks. 20-40 hours per project; 4-12 lessons per track.</em></summary>
  928. <br/>
  929. | # | Project | Combines | Lang |
  930. |:---:|---------|----------|------|
  931. | 01 | [Terminal-Native Coding Agent](phases/19-capstone-projects/01-terminal-native-coding-agent/) | P0 P5 P7 P10 P11 P13 P14 P15 P17 P18 | Python, TypeScript |
  932. | 02 | [RAG over Codebase (Cross-Repo Semantic Search)](phases/19-capstone-projects/02-rag-over-codebase/) | P5 P7 P11 P13 P17 | Python, TypeScript |
  933. | 03 | [Real-Time Voice Assistant (ASR → LLM → TTS)](phases/19-capstone-projects/03-realtime-voice-assistant/) | P6 P7 P11 P13 P14 P17 | Python, TypeScript |
  934. | 04 | [Multimodal Document QA (Vision-First)](phases/19-capstone-projects/04-multimodal-document-qa/) | P4 P5 P7 P11 P12 P17 | Python, TypeScript |
  935. | 05 | [Autonomous Research Agent (AI-Scientist Class)](phases/19-capstone-projects/05-autonomous-research-agent/) | P0 P2 P3 P7 P10 P14 P15 P16 P18 | Python, TypeScript |
  936. | 06 | [DevOps Troubleshooting Agent for Kubernetes](phases/19-capstone-projects/06-devops-troubleshooting-agent/) | P11 P13 P14 P15 P17 P18 | Python, TypeScript |
  937. | 07 | [End-to-End Fine-Tuning Pipeline](phases/19-capstone-projects/07-end-to-end-fine-tuning-pipeline/) | P2 P3 P7 P10 P11 P17 P18 | Python, TypeScript |
  938. | 08 | [Production RAG Chatbot (Regulated Vertical)](phases/19-capstone-projects/08-production-rag-chatbot/) | P5 P7 P11 P12 P17 P18 | Python, TypeScript |
  939. | 09 | [Code Migration Agent (Repo-Level Upgrade)](phases/19-capstone-projects/09-code-migration-agent/) | P5 P7 P11 P13 P14 P15 P17 | Python, TypeScript |
  940. | 10 | [Multi-Agent Software Engineering Team](phases/19-capstone-projects/10-multi-agent-software-team/) | P11 P13 P14 P15 P16 P17 | Python, TypeScript |
  941. | 11 | [LLM Observability & Eval Dashboard](phases/19-capstone-projects/11-llm-observability-dashboard/) | P11 P13 P17 P18 | Python, TypeScript, Go |
  942. | 12 | [Video Understanding Pipeline (Scene → QA)](phases/19-capstone-projects/12-video-understanding-pipeline/) | P4 P6 P7 P11 P12 P17 | Python, TypeScript |
  943. | 13 | [Stateless MCP Server with Registry and Governance](phases/19-capstone-projects/13-mcp-server-with-registry/) | P11 P13 P14 P17 P18 | Python, TypeScript, Go |
  944. | 14 | [Speculative-Decoding Inference Server](phases/19-capstone-projects/14-speculative-decoding-server/) | P3 P7 P10 P17 | Python, TypeScript |
  945. | 15 | [Constitutional Safety Harness + Red-Team Range](phases/19-capstone-projects/15-constitutional-safety-harness/) | P10 P11 P13 P14 P18 | Python, TypeScript |
  946. | 16 | [GitHub Issue-to-PR Autonomous Agent](phases/19-capstone-projects/16-github-issue-to-pr-agent/) | P11 P13 P14 P15 P17 | Python, TypeScript |
  947. | 17 | [Personal AI Tutor (Adaptive, Multimodal)](phases/19-capstone-projects/17-personal-ai-tutor/) | P5 P6 P11 P12 P14 P17 P18 | Python, TypeScript |
  948. **Deep-build tracks** — multi-lesson series that build a complete subsystem from scratch.
  949. | # | Project | Combines | Lang |
  950. |:---:|---------|----------|------|
  951. | 20 | [Agent Harness Loop Contract](phases/19-capstone-projects/20-agent-harness-loop-contract/) | A. Agent harness | Python, TypeScript |
  952. | 21 | [Tool Registry with Schema Validation](phases/19-capstone-projects/21-tool-registry-schema-validation/) | A. Agent harness | Python, TypeScript |
  953. | 22 | [JSON-RPC 2.0 Over Newline-Delimited Stdio](phases/19-capstone-projects/22-jsonrpc-stdio-transport/) | A. Agent harness | Python, TypeScript |
  954. | 23 | [Function Call Dispatcher](phases/19-capstone-projects/23-function-call-dispatcher/) | A. Agent harness | Python, TypeScript |
  955. | 24 | [Plan-Execute Control Flow](phases/19-capstone-projects/24-plan-execute-control-flow/) | A. Agent harness | Python, TypeScript |
  956. | 25 | [Verification Gates and Observation Budget](phases/19-capstone-projects/25-verification-gates-observation-budget/) | A. Agent harness | Python, TypeScript |
  957. | 26 | [Sandbox Runner with Denylist and Path Jail](phases/19-capstone-projects/26-sandbox-runner-denylist/) | A. Agent harness | Python, TypeScript |
  958. | 27 | [Eval Harness with Fixture Tasks](phases/19-capstone-projects/27-eval-harness-fixture-tasks/) | A. Agent harness | Python, TypeScript |
  959. | 28 | [Observability with OTel GenAI Spans and Prometheus Metrics](phases/19-capstone-projects/28-observability-otel-traces/) | A. Agent harness | Python, TypeScript |
  960. | 29 | [End-to-End Coding Agent on the Harness](phases/19-capstone-projects/29-end-to-end-coding-task-demo/) | A. Agent harness | Python, TypeScript |
  961. | 30 | [BPE Tokenizer From Scratch](phases/19-capstone-projects/30-bpe-tokenizer-from-scratch/) | B. NLP LLM | Python, TypeScript, Rust |
  962. | 31 | [Tokenized Dataset with Sliding Window](phases/19-capstone-projects/31-tokenized-dataset-sliding-window/) | B. NLP LLM | Python |
  963. | 32 | [Token and Positional Embeddings](phases/19-capstone-projects/32-token-positional-embeddings/) | B. NLP LLM | Python |
  964. | 33 | [Multi-Head Self-Attention](phases/19-capstone-projects/33-multihead-self-attention/) | B. NLP LLM | Python |
  965. | 34 | [Transformer Block from Scratch](phases/19-capstone-projects/34-transformer-block/) | B. NLP LLM | Python |
  966. | 35 | [GPT Model Assembly](phases/19-capstone-projects/35-gpt-model-assembly/) | B. NLP LLM | Python |
  967. | 36 | [Training Loop and Evaluation](phases/19-capstone-projects/36-training-loop-eval/) | B. NLP LLM | Python |
  968. | 37 | [Loading Pretrained Weights](phases/19-capstone-projects/37-loading-pretrained-weights/) | B. NLP LLM | Python |
  969. | 38 | [Classifier Fine-Tuning by Head Swap](phases/19-capstone-projects/38-classifier-finetuning/) | B. NLP LLM | Python |
  970. | 39 | [Instruction Tuning by Supervised Fine-Tuning](phases/19-capstone-projects/39-instruction-tuning-sft/) | B. NLP LLM | Python |
  971. | 40 | [Direct Preference Optimization from Scratch](phases/19-capstone-projects/40-dpo-from-scratch/) | B. NLP LLM | Python |
  972. | 41 | [Full Evaluation Pipeline](phases/19-capstone-projects/41-eval-pipeline/) | B. NLP LLM | Python |
  973. | 42 | [Large Corpus Downloader](phases/19-capstone-projects/42-large-corpus-downloader/) | C. Train end-to-end | Python, TypeScript |
  974. | 43 | [HDF5 Tokenized Corpus](phases/19-capstone-projects/43-hdf5-tokenized-corpus/) | C. Train end-to-end | Python |
  975. | 44 | [Cosine LR with Linear Warmup](phases/19-capstone-projects/44-cosine-lr-warmup/) | C. Train end-to-end | Python |
  976. | 45 | [Gradient Clipping and Mixed Precision](phases/19-capstone-projects/45-gradient-clipping-amp/) | C. Train end-to-end | Python |
  977. | 46 | [Gradient Accumulation](phases/19-capstone-projects/46-gradient-accumulation/) | C. Train end-to-end | Python |
  978. | 47 | [Checkpoint Save and Resume](phases/19-capstone-projects/47-checkpoint-save-resume/) | C. Train end-to-end | Python |
  979. | 48 | [Distributed Data Parallel and FSDP from Scratch](phases/19-capstone-projects/48-distributed-fsdp-ddp/) | C. Train end-to-end | Python |
  980. | 49 | [Language Model Evaluation Harness](phases/19-capstone-projects/49-lm-eval-harness/) | C. Train end-to-end | Python, TypeScript |
  981. | 50 | [Hypothesis Generator](phases/19-capstone-projects/50-hypothesis-generator/) | D. Auto research | Python, TypeScript |
  982. | 51 | [Literature Retrieval](phases/19-capstone-projects/51-literature-retrieval/) | D. Auto research | Python, TypeScript |
  983. | 52 | [Experiment Runner](phases/19-capstone-projects/52-experiment-runner/) | D. Auto research | Python, TypeScript |
  984. | 53 | [Result Evaluator](phases/19-capstone-projects/53-result-evaluator/) | D. Auto research | Python |
  985. | 54 | [Paper Writer](phases/19-capstone-projects/54-paper-writer/) | D. Auto research | Python, TypeScript |
  986. | 55 | [Critic Loop](phases/19-capstone-projects/55-critic-loop/) | D. Auto research | Python, TypeScript |
  987. | 56 | [Iteration Scheduler](phases/19-capstone-projects/56-iteration-scheduler/) | D. Auto research | Python, TypeScript |
  988. | 57 | [End-to-End Research Demo](phases/19-capstone-projects/57-end-to-end-research-demo/) | D. Auto research | Python, TypeScript |
  989. | 58 | [Vision Encoder Patches](phases/19-capstone-projects/58-vision-encoder-patches/) | E. Multimodal VLM | Python |
  990. | 59 | [Vision Transformer Encoder](phases/19-capstone-projects/59-vit-transformer/) | E. Multimodal VLM | Python |
  991. | 60 | [Projection Layer for Modality Alignment](phases/19-capstone-projects/60-projection-layer-modality-align/) | E. Multimodal VLM | Python |
  992. | 61 | [Cross-Attention Fusion](phases/19-capstone-projects/61-cross-attention-fusion/) | E. Multimodal VLM | Python |
  993. | 62 | [Vision-Language Pretraining](phases/19-capstone-projects/62-vision-language-pretraining/) | E. Multimodal VLM | Python |
  994. | 63 | [Multimodal Evaluation](phases/19-capstone-projects/63-multimodal-eval/) | E. Multimodal VLM | Python |
  995. | 64 | [Chunking Strategies, Compared](phases/19-capstone-projects/64-chunking-strategies-advanced/) | F. Advanced RAG | Python, TypeScript |
  996. | 65 | [Hybrid Retrieval with BM25 and Dense Embeddings](phases/19-capstone-projects/65-hybrid-retrieval-bm25-dense/) | F. Advanced RAG | Python, TypeScript |
  997. | 66 | [Cross-Encoder Reranker](phases/19-capstone-projects/66-reranker-cross-encoder/) | F. Advanced RAG | Python |
  998. | 67 | [Query Rewriting: HyDE, Multi-Query, and Decomposition](phases/19-capstone-projects/67-query-rewriting-hyde/) | F. Advanced RAG | Python, TypeScript |
  999. | 68 | [RAG Evaluation: Precision, Recall, MRR, nDCG, Faithfulness, Answer Relevance](phases/19-capstone-projects/68-rag-eval-precision-recall/) | F. Advanced RAG | Python, TypeScript |
  1000. | 69 | [End-to-End RAG System](phases/19-capstone-projects/69-end-to-end-rag-system/) | F. Advanced RAG | Python |
  1001. | 70 | [Task Spec Format](phases/19-capstone-projects/70-task-spec-format/) | G. Eval framework | Python, TypeScript |
  1002. | 71 | [Classical Metrics](phases/19-capstone-projects/71-classical-metrics/) | G. Eval framework | Python, TypeScript |
  1003. | 72 | [Code Exec Metric](phases/19-capstone-projects/72-code-exec-metric/) | G. Eval framework | Python, TypeScript |
  1004. | 73 | [Perplexity and Calibration](phases/19-capstone-projects/73-perplexity-calibration/) | G. Eval framework | Python, TypeScript |
  1005. | 74 | [Leaderboard Aggregation](phases/19-capstone-projects/74-leaderboard-aggregation/) | G. Eval framework | Python, TypeScript |
  1006. | 75 | [End-to-End Eval Runner](phases/19-capstone-projects/75-end-to-end-eval-runner/) | G. Eval framework | Python, TypeScript |
  1007. | 76 | [Collective Ops From Scratch](phases/19-capstone-projects/76-collective-ops-from-scratch/) | H. Distributed train | Python |
  1008. | 77 | [Data Parallel DDP From Scratch](phases/19-capstone-projects/77-data-parallel-ddp/) | H. Distributed train | Python |
  1009. | 78 | [ZeRO Optimizer State Sharding](phases/19-capstone-projects/78-zero-parameter-sharding/) | H. Distributed train | Python |
  1010. | 79 | [Pipeline Parallel and Bubble Analysis](phases/19-capstone-projects/79-pipeline-parallel/) | H. Distributed train | Python |
  1011. | 80 | [Sharded Checkpoint and Atomic Resume](phases/19-capstone-projects/80-checkpoint-sharded-resume/) | H. Distributed train | Python |
  1012. | 81 | [End-to-End Distributed Training](phases/19-capstone-projects/81-end-to-end-distributed-train/) | H. Distributed train | Python |
  1013. | 82 | [Jailbreak Taxonomy](phases/19-capstone-projects/82-jailbreak-taxonomy/) | I. Safety harness | Python |
  1014. | 83 | [Prompt Injection Detector](phases/19-capstone-projects/83-prompt-injection-detector/) | I. Safety harness | Python, TypeScript |
  1015. | 84 | [Refusal Evaluation](phases/19-capstone-projects/84-refusal-evaluation/) | I. Safety harness | Python |
  1016. | 85 | [Content Classifier Integration](phases/19-capstone-projects/85-content-classifier-integration/) | I. Safety harness | Python, TypeScript |
  1017. | 86 | [Constitutional Rules Engine](phases/19-capstone-projects/86-constitutional-rules-engine/) | I. Safety harness | Python |
  1018. | 87 | [End-to-End Safety Gate](phases/19-capstone-projects/87-end-to-end-safety-gate/) | I. Safety harness | Python, TypeScript |
  1019. </details>
  1020. ## Books and certifications
  1021. <details>
  1022. <summary>Read the core curriculum as a book</summary>
  1023. The 20-phase core curriculum under `phases/` compiles into a six-volume book series. EPUB and PDF are built by CI from the same core lesson sources and attached to every [GitHub release](https://github.com/rohitg00/ai-engineering-from-scratch/releases); the links below always resolve to the newest release. Volume numbers index the series, not versions: each copy carries a dated edition stamp, and older editions stay downloadable from their release.
  1024. Certification curricula are intentionally not converted into the books. Their
  1025. AI tutor state, runnable labs, interactive figures, diagnostics, and timed mocks
  1026. remain first-class on GitHub and the website.
  1027. | Vol | Title | Phases | Download |
  1028. |-----|-------|--------|----------|
  1029. | 1 | Foundations · Math, Tooling, and Classical Machine Learning | 00-02 | [EPUB](https://github.com/rohitg00/ai-engineering-from-scratch/releases/latest/download/aiefs-vol1-foundations.epub) · [PDF](https://github.com/rohitg00/ai-engineering-from-scratch/releases/latest/download/aiefs-vol1-foundations.pdf) |
  1030. | 2 | Deep Learning · Networks, Vision, and Speech | 03, 04, 06 | [EPUB](https://github.com/rohitg00/ai-engineering-from-scratch/releases/latest/download/aiefs-vol2-deep-learning.epub) · [PDF](https://github.com/rohitg00/ai-engineering-from-scratch/releases/latest/download/aiefs-vol2-deep-learning.pdf) |
  1031. | 3 | Language · NLP Foundations and the Transformer | 05, 07 | [EPUB](https://github.com/rohitg00/ai-engineering-from-scratch/releases/latest/download/aiefs-vol3-language.epub) · [PDF](https://github.com/rohitg00/ai-engineering-from-scratch/releases/latest/download/aiefs-vol3-language.pdf) |
  1032. | 4 | Large Language Models · Generation, Reinforcement, Pretraining, and Engineering | 08-11 | [EPUB](https://github.com/rohitg00/ai-engineering-from-scratch/releases/latest/download/aiefs-vol4-llms.epub) · [PDF](https://github.com/rohitg00/ai-engineering-from-scratch/releases/latest/download/aiefs-vol4-llms.pdf) |
  1033. | 5 | Agents · Multimodality, Protocols, Autonomy, and Swarms | 12-16 | [EPUB](https://github.com/rohitg00/ai-engineering-from-scratch/releases/latest/download/aiefs-vol5-agents.epub) · [PDF](https://github.com/rohitg00/ai-engineering-from-scratch/releases/latest/download/aiefs-vol5-agents.pdf) |
  1034. | 6 | Production · Infrastructure, Safety, and Capstones | 17-19 | [EPUB](https://github.com/rohitg00/ai-engineering-from-scratch/releases/latest/download/aiefs-vol6-production.epub) · [PDF](https://github.com/rohitg00/ai-engineering-from-scratch/releases/latest/download/aiefs-vol6-production.pdf) |
  1035. The book is the snapshot; this repository is the living edition. Every chapter ends with links back to the lesson's animated figures, quiz, and runnable code. Build locally with `python3 scripts/build_book.py` (pandoc required); pipeline details in [book/README.md](book/README.md).
  1036. </details>
  1037. <details>
  1038. <summary>Prepare for Claude certifications</summary>
  1039. The [Claude Certification Academy](certifications/claude/README.md) is a free,
  1040. open-source preparation program for all four official Claude certification tracks:
  1041. Associate Foundations, Developer Foundations, Architect Foundations, and Architect
  1042. Professional. Each route combines blueprint-mapped lessons, runnable labs, a
  1043. diagnostic, capstone work, and a full-length original practice exam.
  1044. Use the [AI-native GitHub onboarding guide](certifications/claude/GETTING_STARTED.md)
  1045. with Claude Code, Codex, ChatGPT, Cursor, or another agent. Run
  1046. `claude-certification` in Codex, `/claude-certification` in Claude Code, or ask
  1047. another host to use `claude-certification`. It chooses a track, creates a
  1048. persistent route in `CLAUDE-CERTIFICATION.md`, teaches one step at a time, runs
  1049. the real labs, and gives artifact-based feedback. The same curriculum remains
  1050. available on the [certification website](https://aiengineeringfromscratch.com/certifications.html).
  1051. The academy is independent study material based on public exam objectives. It is not
  1052. affiliated with Anthropic, does not reproduce live exam questions, and cannot guarantee
  1053. a passing score.
  1054. </details>
  1055. <details>
  1056. <summary>Prepare for MCP Associate (MCPA)</summary>
  1057. The [MCPA Certification Curriculum](certifications/mcpa/README.md) is a free,
  1058. open-source preparation program for the Model Context Protocol Associate exam from the
  1059. Agentic AI Foundation, delivered through Linux Foundation Training. Its 34 lessons teach
  1060. the stateless 2026-07-28 protocol across the five exam domains: per-request `_meta` and
  1061. `server/discover` in place of the old handshake, multi round-trip requests, subscriptions,
  1062. caching, the tasks and MCP Apps extensions, OAuth authorization, and the registry and SDK
  1063. tiers. Every lesson ships a runnable standard-library lab whose transcript is checked for
  1064. the current wire shape, and the track adds a diagnostic, a capstone, and three full-length
  1065. original practice exams whose question mix follows the published blueprint weights.
  1066. Use the [AI-native GitHub onboarding guide](certifications/mcpa/GETTING_STARTED.md) with
  1067. Claude Code, Codex, ChatGPT, Cursor, or another agent. Run `mcpa-certification` in Codex,
  1068. `/mcpa-certification` in Claude Code, or ask another host to use `mcpa-certification`. It
  1069. creates a persistent route in `MCPA-CERTIFICATION.md`, teaches one step at a time, runs
  1070. the real labs, and gives artifact-based feedback. The same curriculum is available on the
  1071. [MCPA track page](https://aiengineeringfromscratch.com/certification?id=mcpa-f).
  1072. This curriculum is independent study material based on public exam objectives. It is not
  1073. affiliated with the Agentic AI Foundation or the Linux Foundation, does not reproduce
  1074. live exam questions, and cannot guarantee a passing score.
  1075. </details>
  1076. ## The toolkit
  1077. Every lesson produces a reusable artifact. Install one in your agent or use the scripts below from the repository root.
  1078. <details>
  1079. <summary>Lesson structure and reusable artifacts</summary>
  1080. ## The shape of a lesson
  1081. Each lesson lives in its own folder, with the same structure across the entire curriculum:
  1082. ```text
  1083. phases/<NN>-<phase-name>/<NN>-<lesson-name>/
  1084. ├── code/ runnable implementations (Python, TypeScript, Rust, Julia, Go)
  1085. ├── docs/
  1086. │ └── en.md lesson narrative
  1087. └── outputs/ prompts, skills, agents, or MCP servers this lesson produces
  1088. ```
  1089. Every lesson follows six beats. The *Build It / Use It* split is the spine — you implement the
  1090. algorithm from scratch first, then run the same thing through the production library. You
  1091. understand what the framework is doing because you wrote the smaller version yourself.
  1092. ```mermaid
  1093. %%{init: {'theme':'base','themeVariables':{'primaryColor':'#fafaf5','primaryTextColor':'#1a1a1a','primaryBorderColor':'#3553ff','lineColor':'#3553ff','fontFamily':'JetBrains Mono','fontSize':'13px'}}}%%
  1094. flowchart LR
  1095. M["MOTTO<br/><sub>one-line core idea</sub>"] --> Pr["PROBLEM<br/><sub>concrete pain</sub>"]
  1096. Pr --> C["CONCEPT<br/><sub>diagrams &amp; intuition</sub>"]
  1097. C --> B["BUILD IT<br/><sub>raw math, no frameworks</sub>"]
  1098. B --> U["USE IT<br/><sub>same thing in PyTorch / sklearn</sub>"]
  1099. U --> S["SHIP IT<br/><sub>prompt · skill · agent · MCP</sub>"]
  1100. classDef beat fill:#fafaf5,stroke:#3553ff,color:#1a1a1a
  1101. class M,Pr,C,B,U,S beat
  1102. ```
  1103. ## Every lesson ships something
  1104. Other curricula end with *"congratulations, you learned X."* Each lesson here ends with a
  1105. **reusable tool** you can install or paste into your daily workflow.
  1106. <table>
  1107. <tr>
  1108. <th align="left" width="25%"><img src="site/assets/figures/001-a-prompts.svg" width="120" height="120" alt="FIG_001.A prompts"/><br/><sub>FIG_001 · A</sub><br/><b>PROMPTS</b></th>
  1109. <th align="left" width="25%"><img src="site/assets/figures/001-b-skills.svg" width="120" height="120" alt="FIG_001.B skills"/><br/><sub>FIG_001 · B</sub><br/><b>SKILLS</b></th>
  1110. <th align="left" width="25%"><img src="site/assets/figures/001-c-agents.svg" width="120" height="120" alt="FIG_001.C agents"/><br/><sub>FIG_001 · C</sub><br/><b>AGENTS</b></th>
  1111. <th align="left" width="25%"><img src="site/assets/figures/001-d-mcp-servers.svg" width="120" height="120" alt="FIG_001.D MCP servers"/><br/><sub>FIG_001 · D</sub><br/><b>MCP SERVERS</b></th>
  1112. </tr>
  1113. <tr>
  1114. <td valign="top">Paste into any AI assistant for expert-level help on a narrow task.</td>
  1115. <td valign="top">Drop into Claude, Cursor, Codex, OpenClaw, Hermes, or any agent that reads <code>SKILL.md</code>.</td>
  1116. <td valign="top">Deploy as autonomous workers — you wrote the loop yourself in Phase 14.</td>
  1117. <td valign="top">Plug into any MCP-compatible client. Built end-to-end in Phase 13.</td>
  1118. </tr>
  1119. </table>
  1120. </details>
  1121. <details>
  1122. <summary>Install lesson artifacts</summary>
  1123. **The lesson artifacts.** The repo ships 414 skills and 100 prompts under
  1124. `phases/**/outputs/`; install them via `scripts/install_skills.py`. Requires
  1125. cloning the repo. Supports tag filters, dry-runs, and per-agent layouts:
  1126. ```bash
  1127. python3 scripts/install_skills.py <target> # every skill, default --layout skills (nested)
  1128. python3 scripts/install_skills.py <target> --layout skills # same as above, explicit
  1129. python3 scripts/install_skills.py <target> --type all # skills + prompts + agents
  1130. python3 scripts/install_skills.py <target> --phase 14 # one phase only
  1131. python3 scripts/install_skills.py <target> --tag rag # filter by tag
  1132. python3 scripts/install_skills.py <target> --layout flat # flat files
  1133. python3 scripts/install_skills.py <target> --dry-run # preview without writing
  1134. python3 scripts/install_skills.py <target> --force # overwrite existing files
  1135. ```
  1136. `<target>` is the skills directory for your agent (examples:
  1137. `~/.claude/skills/`, `~/.cursor/skills/`, `~/.config/openclaw/skills/`,
  1138. `.skills/`, or any path your agent reads).
  1139. By default the script refuses to overwrite an existing destination and exits
  1140. with code 1 after listing every colliding path. Use `--dry-run` to preview
  1141. collisions or `--force` to overwrite. Every non-dry-run run writes a
  1142. `manifest.json` in the target with the full inventory grouped by type and
  1143. phase. Pick the layout your agent reads:
  1144. | `--layout` | Path written |
  1145. |---|---|
  1146. | `skills` | `<target>/<name>/SKILL.md` (nested convention, supported by Claude / Cursor / Codex / OpenClaw / Hermes) |
  1147. | `by-phase` | `<target>/phase-NN/<name>.md` |
  1148. | `flat` | `<target>/<name>.md` |
  1149. </details>
  1150. <details>
  1151. <summary>Drop the agent workbench into your own repo</summary>
  1152. The Phase 14 capstone ships a reusable Agent Workbench pack (AGENTS.md, schemas,
  1153. init / verify / handoff scripts). Scaffold it into any repo with:
  1154. ```bash
  1155. python3 scripts/scaffold_workbench.py path/to/your-repo # full pack + seeds
  1156. python3 scripts/scaffold_workbench.py path/to/your-repo --minimal # skip docs/
  1157. python3 scripts/scaffold_workbench.py path/to/your-repo --dry-run # preview only
  1158. python3 scripts/scaffold_workbench.py path/to/your-repo --force # overwrite
  1159. ```
  1160. You get the seven workbench surfaces wired up, a starter `task_board.json`,
  1161. and a fresh `agent_state.json` at `schema_version: 1`. From there: edit the
  1162. task, edit `AGENTS.md`, run `scripts/init_agent.py`, hand the contract to
  1163. your agent. The pack source lives at
  1164. `phases/14-agent-engineering/42-agent-workbench-capstone/outputs/agent-workbench-pack/`.
  1165. </details>
  1166. <details>
  1167. <summary>Browse the entire course as JSON</summary>
  1168. `scripts/build_catalog.py` walks every phase, every lesson, every artifact on
  1169. disk and writes `catalog.json` at the repo root. One file, every course truth.
  1170. ```bash
  1171. python3 scripts/build_catalog.py # writes <repo>/catalog.json
  1172. python3 scripts/build_catalog.py --stdout # to stdout, do not touch repo
  1173. python3 scripts/build_catalog.py --out path/to/file.json
  1174. ```
  1175. The catalog is filesystem-derived, not README-derived, so counts always match
  1176. what is actually on disk. Use it for site builds, downstream tooling, or to
  1177. verify the README counts have not drifted. Schema is documented at the top of
  1178. the script.
  1179. The curriculum workflow builds `catalog.json` as an ephemeral, gitignored
  1180. artifact. Do not commit it. The same workflow runs `audit_lessons.py` as a
  1181. blocking check.
  1182. </details>
  1183. <details>
  1184. <summary>Smoke-check every lesson's Python code</summary>
  1185. `scripts/lesson_run.py` byte-compiles every `.py` file under each lesson's
  1186. `code/` directory. Default mode is syntax-check only — no execution, no API
  1187. keys, no heavy ML deps required. Catches the regressions contributors
  1188. introduce most often (bad indentation, broken f-strings, stray edits).
  1189. ```bash
  1190. python3 scripts/lesson_run.py # syntax-check the whole curriculum
  1191. python3 scripts/lesson_run.py --phase 14 # one phase only
  1192. python3 scripts/lesson_run.py --json # JSON report on stdout
  1193. python3 scripts/lesson_run.py --strict # exit 1 if any lesson fails
  1194. python3 scripts/lesson_run.py --execute # actually run, 10s timeout per lesson
  1195. ```
  1196. `--execute` runs each lesson's `code/main.py` (or the first `.py` file) with a
  1197. 10-second timeout. Lessons whose entry file starts with a `# requires: pkg1,
  1198. pkg2` comment listing non-stdlib deps are skipped with reason `needs <deps>`.
  1199. The script is opt-in and not wired into CI.
  1200. Stdlib only, Python 3.10+. Set `LINK_CHECK_SKIP=domain1,domain2` to override
  1201. the default skip-list (`twitter.com`, `x.com`, `linkedin.com`,
  1202. `instagram.com`, `medium.com` — domains that aggressively block automated
  1203. HEAD/GET).
  1204. </details>
  1205. <details>
  1206. <summary>Foundational papers and protocols</summary>
  1207. - *Attention Is All You Need* — Vaswani et al., 2017 → [Phase 7](#phase-7)
  1208. - *Language Models are Few-Shot Learners* (GPT-3) → [Phase 10](#phase-10)
  1209. - *Denoising Diffusion Probabilistic Models* → [Phase 8](#phase-8)
  1210. - *InstructGPT / RLHF* → [Phase 10](#phase-10)
  1211. - *Direct Preference Optimization* → [Phase 10](#phase-10)
  1212. - *Chain-of-Thought Prompting* → [Phase 11](#phase-11)
  1213. - *ReAct: Reasoning + Acting in LLMs* → [Phase 14](#phase-14)
  1214. - *Model Context Protocol* — Anthropic → [Phase 13](#phase-13)
  1215. </details>
  1216. ## Contributing
  1217. | Goal | Read |
  1218. |---|---|
  1219. | Contribute a lesson or fix | [CONTRIBUTING.md](CONTRIBUTING.md) |
  1220. | Fork for your team or school | [FORKING.md](FORKING.md) |
  1221. | Lesson template | [LESSON_TEMPLATE.md](LESSON_TEMPLATE.md) |
  1222. | Track progress | [ROADMAP.md](ROADMAP.md) |
  1223. | Glossary | [glossary/terms.md](glossary/terms.md) |
  1224. | Code of conduct | [CODE_OF_CONDUCT.md](CODE_OF_CONDUCT.md) |
  1225. Before submitting a lesson, run the invariant check:
  1226. ```bash
  1227. python3 scripts/audit_lessons.py # full curriculum
  1228. python3 scripts/audit_lessons.py --phase 14 # single phase
  1229. python3 scripts/audit_lessons.py --json # CI-friendly output
  1230. ```
  1231. Exit code is non-zero when any rule fails. Rules (L001–L010) validate directory
  1232. shape, `docs/en.md` presence + H1, `code/` non-emptiness, `quiz.json` schema
  1233. (rejects the legacy `q/choices/answer` keys that caused issue #102), and
  1234. relative links inside lesson docs.
  1235. <a id="supporters"></a>
  1236. ## Sponsor the work
  1237. <!-- STATS:START (generated from site/stats.json by build.js — do not edit by hand) -->
  1238. <p align="center"><sub><b>114,584</b> readers &nbsp;·&nbsp; <b>181,995</b> page views in the last 30 days &nbsp;·&nbsp; as of 2026-08-29</sub></p>
  1239. <!-- STATS:END -->
  1240. Free, MIT-licensed, 542 lessons. Thank you to the sponsors and backers who make the work possible.
  1241. [See all sponsors and backers](BACKERS.md).
  1242. Want to support the work? See [sponsorship options](SPONSORS.md), including
  1243. [hardware sponsorships](SPONSORS.md#hardware-lab-partner), or
  1244. [sponsor on GitHub](https://github.com/sponsors/rohitg00).
  1245. If this manual helped you, star the repo. It keeps the project alive.
  1246. ## License
  1247. MIT. Use it however you want — fork it, teach it, sell it, ship it. Attribution appreciated,
  1248. not required.
  1249. Maintained by [Rohit Ghumare](https://github.com/rohitg00) and the community.
  1250. <sub>
  1251. <a href="https://x.com/ghumare64">@ghumare64</a> &nbsp;·&nbsp;
  1252. <a href="https://aiengineeringfromscratch.com">aiengineeringfromscratch.com</a> &nbsp;·&nbsp;
  1253. <a href="https://github.com/rohitg00/ai-engineering-from-scratch/issues/new/choose">Report / Suggest</a>
  1254. </sub>

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.