About

I'm an AI Engineer focused on turning models into production systems that actually work. I build agent workflows, RAG pipelines, and the infrastructure around them—small services, clean APIs, and integrations that slot into existing stacks without friction.

My background is in testing and automation. Years of building test frameworks, hardening CI pipelines, and scaling browser automation taught me to obsess over reliability. That mindset carries into everything I build now: clear contracts, predictable behavior, fast feedback when things break. More on LinkedIn.

Tech Stack

Python is my main language, with TypeScript and Go when needed. I build services with FastAPI, store data in PostgreSQL, use Redis for caching and queues, and ship everything in Docker. For AI work, I reach for LangChain, LlamaIndex, or go direct to APIs—whatever fits the problem. Vector stores for retrieval, self-hosting when it makes sense. I prefer boring, dependable pieces that compose well and stay running.

I also poke around in open source, you can find my work on GitHub. The main one is playwright-distributed, a self-hosted setup for scaling Playwright when you need reliable browser automation at volume.

Get In Touch

If you're building practical AI features or dependable automation and want someone who plugs in quickly, keeps things tidy, and ships steadily, say hi: michal@broton.dev