Newsletter

After the Demo

A newsletter about building AI products that survive contact with the real world. Architectural patterns, production lessons, and honest takes on AI engineering.

What it covers

  • Production architecture for AI systems: context engines, retrieval pipelines, evaluation infrastructure
  • The gap between what works in a demo and what survives real customers
  • Honest assessment of new tooling and techniques as they land
  • Case studies from real product deployments (without the marketing spin)

Sample issues

Context Engines: Building Explainable AI Agents

How to build AI agents that can explain their reasoning — architecture patterns for production explainability.

Introduction to GraphRAG

Graph-based retrieval augmented generation — when flat vector search isn't enough and relationships matter.

Late Chunking and Improved Embeddings

A practical look at why chunking strategy matters more than most teams realize, and what late chunking gets right.

Codebase Modelling

How to build a model of a codebase that an LLM can actually reason over — beyond naive RAG on source files.
See all articles →