Build and optimize low-level, high-performance software and cloud-scale storage systems for multimodal data streams. Focus on OS-level systems programming, performance engineering, and improving data pipelines and I/O for ML workloads. Collaborate with teams to integrate storage formats and ML infrastructure; work onsite in NYC, SF, or London.

Owner of developer-facing content and reference pipelines for ML training data workflows. Produce cookbooks, benchmarks, demos using real datasets, advocate internally for developer needs, answer user questions, and represent the company at conferences and workshops.