Description
Reflection is hiring a Research Program Manager to lead cross-functional programs for training infrastructure and cluster reliability across pre-training, mid-training, and post-training workstreams. The role coordinates with engineering leads and external partners, responds to incidents, improves training-run health and cluster reliability, establishes durable coordination processes, and translates technical complexity into decisions for engineering leadership and executives. Candidates need 7+ years of experience in technical program management, research operations, or infrastructure coordination, ideally in machine learning or large-scale distributed systems, along with deep knowledge of distributed training, GPU clusters, schedulers, networking, and storage.
