Blog

Deep dives into AI engineering, multi-agent systems, and lessons learned building production software in life sciences.

Articles

Minutes of reading

The soundtrack for Birch Aquarium's Imagine Wild exhibit—featuring field recordings of endangered frogs being released into the wild.

A subtle bug in Human-in-the-Loop workflows where edited tool calls leave the agent confused—and how we fixed it with better context engineering

A walkthrough of a production implementation with a runnable example

A deep dive into designing AI-assisted developer workflows by encoding team conventions into a Claude Code skill.

SQL injection, IDOR, agent impersonation, prompt store poisoning, sandboxing, and rollbacks—a practical safety framework for production AI agents.