Member of Technical Staff, Data Engineering

• Build and run data pipelines for large-scale multimodal datasets, from raw video, robotics, and audio through to curated, enriched training data.
• Pull useful signals out of raw data: detecting speakers, isolating background audio, tracking points and features in video, and other signals that affect what the models learn.
• Work on the training mix. Dedup, rebalance clusters, and dig into what's actually in the data so we can find gaps and keep it balanced.
• Work directly with researchers, turning model requirements into a data strategy and their questions into pipeline work.
• Improve the platform: storage, database layers, throughput, and deployment, so the whole system runs faster and more reliably.

What we're looking for

Data is fast becoming one of the biggest bottlenecks in building world models. Our models are only as good as the data behind them, and getting that data right is one of the hardest and most important problems we have. We're looking for a data engineer who wants to be the person figuring it out: building and curating the large-scale multimodal datasets our models train on, across video, robotics, and audio.