Collections / 10 resources
Practitioner Sources
Work from experienced builders, operators, and researchers publishing useful field knowledge.
Reference architecture for context, guardrails, routing, caching, observability and orchestration.
Practical system for context, taste, verification, delegation and closing the loop with AI.
Clear framework for agent tools, planning, failure modes, evaluation and efficiency.
Research hub covering DocETL, EvalGen, agent-first data systems, retrieval debugging and AI data analysts.
Hands-on guide to connecting a custom MCP server to both Claude and ChatGPT web clients.
Current practitioner-oriented guide to choosing AI systems in the agentic era.
Useful perspective on using Claude, ChatGPT and Gemini for different jobs and tracking the open-model ecosystem.
Concrete patterns for evaluating cybersecurity agents using sandboxes, tasks, tools and graders.
Reusable eval skills built from Hamel Husain's work with production AI teams.
Product-oriented argument for making AI outputs inherently easy to verify before layering on evaluation tooling.
