Research Scientist, Real-Time Interactivity / Inference

Highlights

• Invent methods letting people, agents, robots drive world models frame-by-frame
• Shape next-gen real-time interactive video and world model architectures
• Research ships on platform running 15+ fastest world models at scale

What they offer

• Room to influence direction and lead your own research agenda
• Access to real workloads across multiple state-of-the-art model families
• Close collaboration with world-class ML inference engineering team
• Sufficient compute to train and serve models at scale
• Dedicated support for publishing at top conferences
• Competitive SF salary and meaningful early equity
• Visa sponsorship and relocation support
• Generous health, dental, and vision coverage

Ideal candidate

Systems-minded researcher who treats latency, memory, and quality as one problem and wants inventions to ship in real products. You value open, high-impact work and close collaboration with engineers and model builders.

Requirements

• PhD in ML, computer vision, graphics, robotics, or related field, or equivalent practical experience
• Track record in real-time or streaming video generation, autoregressive/causal diffusion, distillation, efficient models, or controllable generation
• Think of latency, memory, and model quality as one problem rather than three
• Fluent in PyTorch and/or JAX and comfortable with large-scale training and inference infrastructure
• Real systems intuition - can read a profiler, reason about memory bandwidth, and work with kernel engineers
• Value quality over quantity in publishing and widely adopted open-source work