ML infrastructure engineer interested in LLM systems, distributed data processing, performance engineering, and low-level systems work.
I spend most of my time around ML infrastructure and LLM data systems. Although most of the time I spent until now has been in CPU-land, I am gradually moving deeper into GPU/model performance profiling.
I also have a soft spot for embedded systems, robotics, Rust, C/C++. Basically anything low-level.
ML infrastructure LLM preprocessing, data pipelines, inference workloads
Distributed processing Ray, DuckDB, Polars, GCP, spot/preemptible infrastructure
Performance work Profiling, bottleneck hunting, CPU/GPU efficiency