AI Engineering

AI Engineering from First Principles

Go beyond the API call. Understand embeddings, retrieval, grounded generation, and the engineering behind reliable AI systems.

Intermediate
AIRAGC#

Build an understandable AI system

An application around a language model is still a software system. It needs explicit boundaries, measurable behavior, authorization, tests, and an honest account of uncertainty.

A practical learning sequence

  1. Represent text with embeddings and compare vectors.
  2. Separate offline indexing from online retrieval.
  3. Build a grounded prompt using authorized sources.
  4. Evaluate retrieval and answers independently.

The lessons connect to the RAG lab. You do not need a paid model account to study the architecture or run the deterministic vector examples.

Lessons in this path

  1. 01Understanding RAG Architecture from First Principles24 min
  2. 02Understanding Embeddings and Vector Similarity20 min

// the notebook, by email

Learn alongside me.

Engineering ideas worth understanding.