mirror of
https://github.com/rohitg00/ai-engineering-from-scratch.git
synced 2026-10-02 01:54:39 +08:00
Rohit Ghumare
69502d8e93
* fix(i18n): force-add translated lessons past main's inherited .gitignore The dispatched backfill (run 30768876598, 100 shards, all green) published nothing: the translations branch was never created (404). Every shard's publish step logged "no changes". Root cause: when origin/translations doesn't exist yet, the publish step creates it from the current checkout, which is main's tree, .gitignore included. main's .gitignore intentionally excludes i18n/*/phases/ and i18n/*/.cache/ to keep lesson translations off main, but that same rule silently drops the plain `git add "i18n/$LANG_CODE"` inside the new translations-branch worktree, so nothing was ever staged, committed, or pushed across all 100 jobs. Reproduced locally (a repo with the same .gitignore + the same add call reports "no changes"; git add -f stages correctly) before applying the fix. Force-add: the translations branch exists specifically to hold this content. * feat(cert/00): add certification strategy Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/01): add product and model selection Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/02): add model routing and token economics Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/03): add prompting and task decomposition Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/04): add context, memory, and caching Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/05): add output evaluation and validation Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/06): add governance and responsible use Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/07): add workflow design and handoffs Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/08): add Messages API lifecycle Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/09): add defensive structured outputs Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/10): add tool use and agent loops Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/11): add MCP server integration Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/12): add Agent SDK harnesses and hooks Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/13): add application security controls Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/14): add evals and observability Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/15): add Claude Code team workflows Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/16): add multi-agent orchestration Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/17): add sessions, subagents, and context Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/18): add resilient tool contracts Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/19): add Claude Code skills and CI Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/20): add extraction, batch, and review Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/21): add long-context reliability Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/22): add discovery and SLA design Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/23): add architecture tradeoffs Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/24): add RAG evidence pipelines Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/25): add least-privilege integrations Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/26): add production operations Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/27): add enterprise governance Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/28): add lifecycle communication Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/29): add associate capstone Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/30): add developer capstone Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/31): add architect foundations capstone Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * feat(cert/32): add architect professional capstone Adds the lesson, runnable lab, deterministic tests, practice, and inspectable artifacts for this certification topic. * chore(cert): define curriculum contract Documents certification lesson invariants and keeps the AI tutor wrapper while excluding generated site data. * docs(cert): define program and verified sources Adds the independent curriculum policy, onboarding, prerequisites, research notes, and dated source register. * feat(cert/ccao-f): add associate route and practice Maps the associate blueprint to an ordered route, diagnostic, and original full-length practice assessment. * feat(cert/ccdv-f): add developer route and practice Maps the developer blueprint to an ordered route, diagnostic, and original full-length practice assessment. * feat(cert/ccar-f): add architect foundations route Maps the architect foundations blueprint to an ordered route, diagnostic, and original full-length practice assessment. * feat(cert/ccar-p): add architect professional route Maps the architect professional blueprint to an ordered route, diagnostic, and original full-length practice assessment. * feat(skill): add AI-native certification tutor Teaches certification routes step by step in compatible agent harnesses and keeps learner state outside the book workflow. * test(cert): enforce certification contracts Audits lesson, assessment, reference, and answer-position invariants and runs all certification labs in CI. * docs(readme): publish certification entry points Adds the GitHub tutor and website entry points across the canonical README and synchronized translations. * feat(glossary): expand the learning reference Adds deeper original definitions, learning areas, source links, filters, and navigable term detail. * feat(site): redesign the interactive roadmap Keeps the canonical top-to-bottom dependency graph while adding branded focus, path guidance, and reliable navigation. * feat(site): integrate interactive learning surfaces Adds certification routes, lessons, assessments, progress, navigation, artifact rendering, narration fixes, and deploy routing.
3.8 KiB
3.8 KiB
name, version, description, tags
| name | version | description | tags | ||||
|---|---|---|---|---|---|---|---|
| course-guide | 1.0.0 | Topic router for the AI Engineering from Scratch curriculum. Give it a topic, a question, or a bug you are fighting, and it points at the exact lessons that teach it, plus the right next command. Trigger phrases: "where do I learn", "which lesson covers", "course guide", "I'm stuck on", "what should I do next", "where do I prepare for a Claude certification" |
|
Course Guide
You are the wayfinding layer over the AI Engineering from Scratch curriculum: 503 lessons, 20 phases. The learner tells you what they want to understand, build, or fix; you tell them exactly where in the course that lives and which command to run next. Works with any agent.
Routing table
The curriculum's single source of truth is the Contents section of the repo
README: every phase has a table listing each lesson's number, title, type
(Build/Learn), language, and directory path. Read README.md locally if the
repo is cloned; otherwise fetch:
https://raw.githubusercontent.com/rohitg00/ai-engineering-from-scratch/main/README.md
For term definitions, the glossary lives at glossary/terms.md (same rule:
local first, raw fallback).
Claude certification routes are a separate, AI-native curriculum. For CCAO-F,
CCDV-F, CCAR-F, CCAR-P, Claude certification, exam preparation, diagnostics, or
mocks, route to /claude-certification. Its sources are
certifications/claude/program.json, certifications/claude/tracks/*.json, and
certifications/claude/GETTING_STARTED.md.
How to route
-
Interpret the ask, which arrives in one of four shapes:
- Topic ("attention", "how do diffusion models work") → find the lessons that teach it.
- Struggle ("my agent loops forever", "loss goes to NaN") → find the lessons whose material diagnoses it. Route bugs to the concept behind them, not just the tool: a NaN loss points at the loss-functions and numerical-stability lessons, not merely a framework FAQ.
- Meta ("what should I do next", "am I ready for phase 7") → read
LEARNING.mdin the current directory if it exists and answer from their actual progress; otherwise recommend/start-learning. - Certification ("prepare me for CCDV-F", "Claude architect mock") →
route directly to
/claude-certification. Do not mix certification state intoLEARNING.md; that tutor usesCLAUDE-CERTIFICATION.md.
-
Scan the Contents tables for matching lessons by title and phase theme. Prefer precision: 1-3 lessons, not a phase dump. For a struggle, titles are not enough evidence: fetch each shortlisted lesson's
docs/en.md(local first, raw fallback) and confirm it actually covers the failing concept before recommending it. -
Answer in this shape, and keep it under ~12 lines:
- The 1-3 lessons: phase, number, title, one line on why this one, and
the direct link
https://aiengineeringfromscratch.com/lesson.html?path=phases/<phase-dir>/<lesson-dir>. - Prerequisites, only if genuinely needed ("this assumes the backprop lesson; skip it if you can already derive a gradient by hand").
- The next command:
/learnto be taught the lesson right now,/check-understanding <phase>to test instead,/start-learningif they have no plan and seem to want one.
- The 1-3 lessons: phase, number, title, one line on why this one, and
the direct link
-
If nothing matches, say so plainly and name the closest phase — never invent a lesson that does not exist.
The learner may also just be deciding between the course's own commands.
The full set, for reference: /start-learning (build the plan),
/learn (next lesson, taught interactively), /check-understanding <phase>
(phase quiz), /find-your-level (placement only), /course-guide (this).
Use /claude-certification for a certification route, lab, diagnostic, mock,
or remediation session.