Arthur Katcher

Python, TypeScript, Claude Code, OpenAI Agents SDK, MCP, AWS Bedrock, Docker, FastAPI, Postgres, Redis, Next.js

How to measure computer use agents and the benchmarks behind them.

Agent harnesses, computer use evals and agent security hooks.

Advisory and builds on agent harnesses, runtimes and deployment. DMs open on X, or book a call at arthurkatcher.com/calendar