Sean Brar

I’m an ML systems engineer and Google DeepMind GSoC 2025 alumnus.

I build evaluation and verification infrastructure for AI systems: harnesses, validation pipelines, and correctness guarantees that determine whether model outputs can be trusted. My GSoC project, Pollux, is an async library for multimodal LLM orchestration. It cuts API costs by 90% through content-hash caching and single-flight deduplication.

Background: BA in Political Science (summa cum laude) from Cal State LA, Caltech AI/ML Certificate, systems engineering at ELEMENTS. Currently pursuing post-baccalaureate CS and Mathematics, preparing for graduate research.

Projects: Pollux · ContextRAG · gh-templates · paperweight

Research interests: ML evaluation and verification, reinforcement learning, and reproducibility infrastructure.