Talks & publications
- 2026Kotlin Server-side in Google SearchKotlinConf 2026
- 2013Scheduling Sensors for Monitoring Sentient Spaces using an Approximate POMDP PolicyIEEE PerCom 2013 · extended in Pervasive and Mobile Computing
I'm a computer engineer, the author of Deep RL: Zero to Hero, a husband and father, and currently at Google, building the infrastructure behind AI in Search. I also write about physics, software, and things I learn along the way.
I studied computer engineering at the University of Bologna and did research on context-aware sensor systems at UC Irvine. After building banking and e-commerce systems in Italy, I moved to San Francisco to join ThousandEyes right after its Series A. Since 2017 I've been at Google, where I design and lead large-scale infrastructure, most recently for AI in Search.
Throughout, my work has sat where distributed systems meet programming languages: runtimes, declarative data flows, and the plumbing that makes software fast and correct at scale.
drlzh.ai is a hands-on reinforcement learning course: 18 notebooks where you implement every algorithm yourself, from MDPs and DQN to PPO, AlphaZero, RLHF and world models, with an AI tutor that lives in VS Code. Over 2k stars on GitHub.
The first two editions are entirely human-authored; only the third was built with AI assistance.
I work on the serving infrastructure behind AI Overviews and AI Mode in Google Search, and lately on the foundations for agents in Search. I spoke about our Kotlin server-side stack at KotlinConf 2026.
Outside of work I'm a husband and a father, which is the most important job I have and the one I'm still learning every day.