AI From Zero to Builder
Resources
Setup commands and canonical external references used by the curriculum, kept separate from lesson prose so learners can find the underlying tools, papers, specifications, and safety guidance.
Run the learning site
Website development supports Node.js 20 or newer. Repository-level validation uses Node.js 22.
cd website
npm ci
npm run startBefore submitting a website change, run the shared website regression suite and production build:
cd website
npm run test:wave1
npm run buildUnder Node.js 22, curriculum metadata can be checked from the repository root with node curriculum/validate.mjs. Browser-Lab-specific smoke tests live under labs/browser.
Version-sensitive reference pins
These records are generated from curriculum/versions.yaml and resolve back to the canonical reference library.
- A2A specification · 1.0.0Reviewed 2026-09-29 · Agent2Agent Protocol — v1.0.0 Specification
- MCP specification · 2026-07-28Reviewed 2026-09-29 · Model Context Protocol — 2026-07-28 Specification
- OpenTelemetry specification · reviewed-2026-09-08Reviewed 2026-09-08 · OpenTelemetry — Specifications
- OWASP LLM Top 10 · reviewed-2026-09-08Reviewed 2026-09-08 · OWASP GenAI Security Project — Top 10 for LLM Applications
- PEFT docs · reviewed-2026-09-08Reviewed 2026-09-08 · Hugging Face — PEFT Documentation
- vLLM docs · reviewed-2026-09-08Reviewed 2026-09-08 · vLLM — Documentation
Core canonical references
A compact cross-section of the reference registry covering ML foundations, Transformers, model adaptation, RAG, agents, observability, risk management, and application security.
- Google for Developers — Introduction to Machine Learning
- scikit-learn — Getting Started
- Vaswani et al. — Attention Is All You Need
- Kudo and Richardson — SentencePiece
- Radford et al. — Language Models are Unsupervised Multitask Learners
- Hu et al. — LoRA: Low-Rank Adaptation of Large Language Models
- Lewis et al. — Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks
- Yao et al. — ReAct: Synergizing Reasoning and Acting in Language Models
- OpenTelemetry — SpecificationsCanonical reference · reviewed 2026-09-08
- NIST AI Risk Management Framework (AI RMF 1.0)
- NIST AI 600-1 — Generative Artificial Intelligence Profile
- OWASP GenAI Security Project — Top 10 for LLM ApplicationsCanonical reference · reviewed 2026-09-08
The generated library currently contains 101 canonical references. Individual lessons link the references relevant to their own topic rather than treating this page as a substitute for lesson-specific evidence.