How monhelpierre/ai-engineering-from-scratch works
This repository is a hands-on AI engineering curriculum, not a single application: learners progress from foundations through model engineering and agents, building runnable lesson artifacts along the way. The documented lesson loop moves from concepts to a from-scratch implementation, then production-library use and a reusable deliverable. Learners can browse the curriculum on the site or use documented placement and review skills. The tree and sampled source support the site and several lesson examples; most lesson-to-lesson runtime wiring is not shown, so the graph represents documented curriculum progression rather than inferred code calls.
Main entry points: 03-deep-learning-core Deep learning core · 10-llms-from-scratch LLMs from scratch · 14-agent-engineering Agent engineering · 19-capstone-projects Capstone projects
Components 20
Learning experience
- Curriculum site
site/app.js - Placement skills
- Lesson workflow
- Reusable artifacts
Technical foundations
- Math foundations
phases/01-math-foundations - ML fundamentals
phases/02-ml-fundamentals
Models and modalities
- Deep learning core
phases/03-deep-learning-core - Computer vision
phases/04-computer-vision - NLP and audio
phases/05-nlp-foundations-to-advanced - Transformers and GenAI
phases/07-transformers-deep-dive
LLM engineering
- LLMs from scratch
phases/10-llms-from-scratch - LLM engineering
phases/11-llm-engineering - Multimodal AI
phases/12-multimodal-ai - Tools and protocols
phases/13-tools-and-protocols
Agents and projects
- Agent engineering
phases/14-agent-engineering - Autonomous systems
phases/15-autonomous-systems - Capstone projects
phases/19-capstone-projects - Coding-agent demo
phases/19-capstone-projects/29-end-to-end-coding-task-demo/code/main.py - End-to-end RAG
phases/19-capstone-projects/69-end-to-end-rag-system/code/main.py
Other components
- Learner
Connections 20
- Learner → Curriculum site: browses
- Curriculum site → Learner: shows lessons
- Learner → Placement skills: requests guidance
- Placement skills → Learner: suggests starting level
- Learner → Lesson workflow: studies and builds
- Lesson workflow → Reusable artifacts: produces
- Math foundations → ML fundamentals: prepares for
- ML fundamentals → Deep learning core: prepares for
- Deep learning core → Computer vision: supports
- Deep learning core → NLP and audio: supports
- Deep learning core → Transformers and GenAI: supports
- Transformers and GenAI → LLMs from scratch: prepares for
- LLMs from scratch → LLM engineering: prepares for
- LLMs from scratch → Multimodal AI: prepares for
- LLM engineering → Tools and protocols: prepares for
- Tools and protocols → Agent engineering: prepares for
- Agent engineering → Autonomous systems: extends to
- Agent engineering → Capstone projects: applies in
- Capstone projects → Coding-agent demo: includes
- Capstone projects → End-to-end RAG: includes