Reading List

The essays, papers, and posts behind the knowledge bank. Grouped by the tier they most support. Add entries as you read.

Foundational essays

• Anthropic — Building effective agents — the essay on workflows vs agents and the building-block patterns. Start here.
• Anthropic — Effective context engineering for AI agents — curating the context window; the #1 applied skill.
• Anthropic — Effective harnesses for long-running agents — reliability, checkpoints, failure recovery.

Evals

• Hamel Husain — Your AI Product Needs Evals — the case for evals and how to start.
• Hamel Husain — LLM-as-a-Judge — scoring outputs at scale with a rubric.

RAG & LLM systems

• Eugene Yan — Patterns for Building LLM-based Systems & Products — a practitioner's map of the patterns.
• pgvector — the committed vector store.

Architecture & multi-agent (the debate)

• Cognition — Don't Build Multi-Agents — the case against speculative multi-agent complexity.
• Cognition — Multi-Agents: What's Actually Working — the nuanced follow-up.
• Read these against Anthropic's Building effective agents — the disagreement is the lesson.

Production & tooling

• Anthropic — Prompt engineering overview · Prompt caching · Tool use
• OpenAI — Structured Outputs
• Langfuse — LLM observability docs

Books & broader reference

• Chip Huyen — AI Engineering (O'Reilly) · aie-book resources
• Simon Willison's blog — frequent, practical, current.