DEV Community

Saurabh Pathak
Saurabh Pathak

Posted on

I built a free 132-topic AI Engineer Roadmap with runnable code (open source)

"Learn AI engineering" usually means a dozen browser tabs and no order. I wanted one structured path, so I built it and made it open source.

Repo: https://github.com/techsaurabh-394/ai-engineer-roadmap
Live site: https://mernstackdev.com/ai-roadmap/

What is in it

132 topics across 10 phases (0 to 9): ML foundations, RAG, advanced RAG, LLM concepts, agents, frameworks, MCP, production AI, fine-tuning and multimodal.

Every topic has:

  • a plain-English explanation with an everyday analogy
  • a diagram
  • Python code
  • 3 interview questions
  • links to the original papers or official docs

Try it in 5 minutes

git clone https://github.com/techsaurabh-394/ai-engineer-roadmap.git
cd ai-engineer-roadmap
python3 -m venv .venv && source .venv/bin/activate
pip install -r requirements.txt
pytest -q
python 01-rag-foundations/code/mini_rag.py "Why use multi-stage builds?"
Enter fullscreen mode Exit fullscreen mode

The last command runs a tiny but complete RAG pipeline offline: it chunks three sample documents, scores them against your question and prints the best passage with a citation. No API key is needed. If you set ANTHROPIC_API_KEY, Claude writes the answer from the retrieved passages instead.

Pick your path

You are Start with
Complete beginner Start here, then "What is RAG?"
Know Python, new to AI What is RAG, then What is an AI Agent
Building agents or MCP servers Agents, then Model Context Protocol
Interview in a week The interview-prep folder

Extras

  • an animation for every topic, plus an interactive viewer
  • flash-card quizzes built from the checkpoint questions; progress stays in your browser
  • CI on the repo, so the examples are tested

What it is not

  • It is not a replacement for building real projects. Reading is step one.
  • Some extra snippets need packages or API keys; the table in the README says which.
  • It is a one-person project, so expect rough edges and some mistakes.

Help me fix it

If you find a wrong explanation, a broken example or a missing topic, open an issue or a PR. Contributions are welcome and I review them properly.

Which topic would you want explained better? Tell me and I will prioritise it.

Top comments (0)