Research Scientist - Data and State Acceleration - Global Frontier Tech Recruitment Program - 2027 Start (PhD)

Build real-time and offline data architectures for large-scale recommendation systems, powering feature pipelines, model training, and real-time inference. Work on streaming Lakehouse systems and unified infrastructure that connects training data with training/inference state for multimodal foundation models. Partner with ML platform teams to support PyTorch training workflows, optimize data formats/access patterns, and own core distributed storage and processing components including file formats, compaction, and metadata management.