Generative AI Engineering
Building products on top of language models: retrieval, evaluation, cost, safety, and everything that decides whether the demo survives real users.
Make this roadmap your plan
Sign in to save your routine, streaks, notes, and progress across devices.
Foundations
The minimum you need to reason about a model rather than prompt it superstitiously.
Python for AI work
beginner1h 30mTyped, testable Python with async I/O and virtual environments. Most AI engineering is ordinary engineering around a network call.
How a language model produces text
beginner1h 30mEmbeddings and vector similarity
beginner1h 15mChoosing a model
beginner1hPrompting and structured output
Getting reliable, parseable behaviour out of a probabilistic system.
Prompt and context engineering
beginner1h 30mStructured output
intermediate1h 15mTool use and function calling
intermediate1h 30mTreating prompts as code
intermediate1hRetrieval-augmented generation
The default architecture for grounding a model in your data — and the one most often built badly.
Chunking and ingestion
intermediate1h 15mVector stores and indexes
intermediate1h 30mHybrid search and reranking
advanced1h 30mGrounding and citations
advanced1h 15mEvaluating a RAG system
advanced1h 30mEvaluation and observability
The discipline that separates a product from a demo.
Building an evaluation set
intermediate1h 30mLLM-as-judge, and its limits
advanced1h 15mRegression testing prompts and models
advanced1h 15mObservability for LLM features
advanced1h 15mProduction concerns
Cost, latency, safety and the failure modes that only appear with real users.
Cost and latency engineering
advanced1h 30mReliability around a probabilistic dependency
advanced1h 15mSafety and abuse resistance
advanced1h 30mPrivacy and data handling
advanced1hGoing further
The techniques worth reaching for once retrieval and evaluation are solid.