Interesting finds, updated throughout the week. Showing latest 3 additions.
Cursor on how Bugbot learns from feedback to improve code review over time — a concrete self-improving-agent loop in production.
Mitchell Hashimoto's first-hand account of integrating AI coding tools into real engineering work. Grounded, no hype.
browser-use applies Sutton's bitter lesson to agent harness design — stop hand-engineering, let general methods scale. Provocative and concrete.
LLM-generated research, prompt explorations, and AI-assisted deep dives.
A deep dive comparing open-source and closed-source AI memory/knowledge graph solutions (vector DBs, graph DBs, hybrid) for agentic workflows, evaluating features like retrieval style, scalability, and LLM optimization.
A comprehensive analysis of leading open-source LLM observability, evaluation, and testing platforms, including pros, cons, intended audience, hosting requirements, and a comparison table.
A deep research exploration of AI agent patterns, their evaluation, and comparison for solutions architects.
Experimental projects and learning explorations in AI and development
Detailed tutorials and best practices for AI development and implementation
Brewing the perfect guides...
Just need one more cup of coffee to finish writing these.
"Welcome to my digital garden — a delightful mix of carefully crafted guides and my somewhat chaotic (but hopefully useful) learning notes. It's like a cookbook where half the recipes are tested, and half are my excited 3 AM experiments."
— David