Read the latest
Home/Guides
Guides
3 step-by-step series, 66 parts, all free to read. Each part has runnable code, diagrams and the pitfalls that matter in production.
AI Engineering
LLMs, RAG, agents, MCP, observability and guardrails: how to build AI features that hold up in production, with runnable Python.
- What an LLM actually is
- Tokens, embeddings & the transformer
- Sampling, latency & cost
- Prompt engineering that holds up
- Why RAG exists
- Embeddings & vector databases
OpenAI SDK
The OpenAI Python SDK end to end: Responses vs Chat, streaming, structured outputs, tools, embeddings, retries, cost and evals.
- How the OpenAI API works
- Setup, keys & your first call
- Responses API vs Chat Completions
- Models, parameters & tokens
- Streaming responses
- Structured outputs with Pydantic
FastAPI
Production FastAPI: parameters, Pydantic, dependencies, async, databases, auth, testing, performance and deployment.
- What FastAPI really is
- Setup, tooling & the dev loop
- Path, query & header parameters
- Request bodies & Pydantic v2
- Responses & serialization
- Dependency injection in depth