About

I work on the Data Infra team at xAI and X. We’re responsible for the CPU and GPU fleets that power xAI data processing, supporting Grok pretraining/midtraining/posttraining data, supervised fine-tuning (SFT), data analysis for Grok Build and Grok Bot, and more. We also run data infra at petabyte-scale for X (Twitter), including ads, recommendation systems, backend core services, and product analytics.

Before xAI, I worked on Tesla’s Cell Manufacturing team, where I led a team that built software to enable more efficient manufacturing operations. My team ran the data platform reponsible for factory reporting and financial tracking.

My technical strengths are distributed data processing engines (Spark, Ray, Flink), analytical data platforms (data lakes, Iceberg, ClickHouse), data pipelines and analytics, and data verifiability.

I studied Chemical Engineering at UC Berkeley. During my degree, I researched high-capacity, nickel-rich battery cathodes at Berkeley Lab and Tesla.

Outside of work, I enjoy learning about functional programming, programming languages, type systems, testing theory/formal methods, and math. I’ve been working on a project to rebuild Spark from scratch in functional Scala 3 with acceleration from vectorization/SIMD.

This site uses GoatCounter for privacy-friendly analytics. It does not use cookies or collect personal data.