Technical Lead Manager - Training Runtime, Data(set) Movement

Highlights

• Build unified dataset read platform powering frontier runs like GPT-5.5.
• Own fast, reliable in-cluster data movement for large-scale training.
• Build visualizers to inspect text, multimodal and RL data in pipeline.

What they offer

• Equity

Ideal candidate

Hands-on distributed systems leader who obsesses over developer experience and correctness at scale. You lead through code first, thriving on making massive training data fast, observable, and reliable for researchers.

Requirements

• Built or owned dataset, data loading, storage, or distributed training infrastructure at large scale
• Care equally about API design, debugging ergonomics, performance, and bit-level correctness
• Understand failure modes of large distributed training jobs and how data systems create or prevent them
• Experience with stateful iterators, checkpoint/restart semantics, caching, remote services, or high-throughput storage reads
• Comfortable working across Python and lower-level systems code
• Worked with multimodal, video, reinforcement learning, or pretraining data pipelines
• Can lead through code before a team exists, and later manage engineers hands-on
• Obsess over developer experience and eliminating friction for researchers