I work on the quality and cost of AI agents in software development.
I'm a Software Engineer at Allegro, working in the JVM ecosystem. I show teams how to check whether code written with AI agents is good enough and how much it really costs. I run trainings and speak at conferences about architecture, refactoring and system design.
Before I became a programmer, I was a civil engineer. That's where my habit comes from: I look at code together with the business and the organization that maintains it.
- cezarysanecki.pl 🇵🇱 PL — trainings, talks, blog
- Ponad Kodem 🇵🇱 PL — code, architecture, business and AI for programmers
- YouTube 🇵🇱 PL
- Agent eval gate — a Langfuse-traced eval that blocks AI agent prompt regressions before push and in CI
- Parking domain — domain modeling of a parking lot, 100-commit contest
- System Design in Practice — conference talk deck
- Refactoring playground — code from my refactoring and unit testing talks and workshops, a base for refactoring practice with or without AI
- Local Whisper — native macOS app for offline Polish speech transcription
- Talks slides — slides from my conference talks
Looking for a training or a talk on the quality and cost of AI agents? Message me on LinkedIn.
