epochs

Paths · 9 modules · 50 lessons · 6 projects · about 24 hours

AI Engineer

Build products on top of language models: prompts, retrieval, agents, evaluation and running it all in production.

Who it's for

Software developers who want to build with LLMs. No ML background needed.

By the end, you can
  • Explain how LLMs turn text into answers, and where they fail
  • Build retrieval-augmented and agent systems that cite sources and stay within limits
  • Measure quality, cost and latency, and keep them in check in production

The route

Modules in order, with a project after the lessons it builds on. Mark lessons as done and tick project steps, and this page tracks your progress. Sign in to keep it on every device.

  1. Module 05 · 4 lessons · about 18 minTokens and embeddingsHow text becomes numbers a model can work with.Not started
  2. Module 06 · 6 lessons · about 26 minTransformersAttention, the full architecture, and how the next token is chosen.Not started
  3. Project · Beginner · 2–3 hours · FreeA tiny language modelGenerate new names by predicting one letter at a time, the same idea as an LLM predicting tokens.4 steps
  4. Module 08 · 6 lessons · about 24 minPrompting and contextPrompts, structured outputs, tools, and what to put in the window.Not started
  5. Module 09 · 5 lessons · about 21 minSearch and retrievalRAG from chunking to reranking, and how to evaluate it.Not started
  6. Project · Beginner · 2–3 hours · ProA search engine for your notesBuild BM25, the keyword search behind most search boxes, and point it at your own notes.5 steps
  7. Project · Intermediate · 4–6 hours · ProChat with your documentsBuild retrieval-augmented generation end to end: chunking, retrieval, a grounded prompt, citations and an evaluation.6 steps
  8. Module 11 · 5 lessons · about 16 minEvaluationDecide what good means, then measure it without fooling yourself.Not started
  9. Project · Intermediate · 3–4 hours · ProAn evaluation harnessBuild the tool every AI team needs: run a fixed test set, see what passed, and know whether a change really helped.5 steps
  10. Module 10 · 6 lessons · about 20 minAgentsThe agent loop, memory, tools, and how agents fail.Not started
  11. Project · Advanced · 4–6 hours · ProA tool-using agentBuild the agent loop yourself: tools, a step limit, a trace of every action, and tests that don't need an API.5 steps
  12. Module 12 · 6 lessons · about 19 minInference and servingWhere the time goes, and how to make models fast and cheap.Not started
  13. Module 13 · 7 lessons · about 21 minProduction AI and MLOpsRouting, fallbacks, budgets, monitoring and safety in real systems.Not started
  14. Project · Intermediate · 2–3 hours · ProPut your project onlineTurn the notes search engine into a tested web API, package it with Docker, and give it a public address.6 steps
  15. Module 14 · 5 lessons · about 16 minML system designA repeatable way to design recommenders, search and LLM products.Not started