NVIDIA's OmniDreams and FlashDreams are the foundation for a new generation of interactive world-model systems. With this release, NVIDIA has set the standard for fidelity, real-time inference performance in world models. And, we've put our work in the open with an open weights model release with an associated open source inferencing engine. We are on a mission to keep raising that bar, making interactive world models viable wherever dynamic worlds matter: autonomous vehicles, robotics, rendering, and simulation.

We are seeking an engineering leader to transform this technology lead into an industry standard platform and ecosystem. You will form the team that keeps FlashDreams fast, makes OmniDreams useful in real workflows, and helps customers and researchers succeed with the stack. Success means more than shipping software: it means high-quality solutions, trusted releases, strong performance, engaged developers, and research feedback that compounds. You'll work alongside outstanding research teams and NVIDIA leadership. If turning breakthrough research into something developers rely on every day sounds like the best job you can envision, we'd love to talk.

What you'll be doing:

• Scale the engineering team building NVIDIA's interactive world-model platform across OmniDreams, FlashDreams, AlpaSim technologies.

• Establish the production engineering path for interactive world models in AV, rendering, robotics, and simulation.

• Build the architecture and processes needed to turn fast paced research into reliable developer-facing software.

• Own product and engineering excellence: performance, scalability, reliability, developer experience, CI/CD, and release management.

• Partner with research, simulation, rendering, and AV teams to translate world-model breakthroughs into production-ready platform capabilities.

• Grow the open developer and research ecosystem around NVIDIA's world-model stack.

What we need to see:

• Demonstrated research or applied innovation in world models, video diffusion, neural rendering, Gaussian splatting / NeRFs, neural reconstruction, neural shading, simulation, robotics, or AV.

• Technical fluency in the ML primitives behind interactive world models, including diffusion or flow-matching models, autoregressive / causal video generation, self-forcing or causal-forcing style training, Gaussian splatting, NeRFs, and neural reconstruction.

• Demonstrated experience turning research-stage systems into dependable, scalable products.

• 12+ overall years experience, 5+ years leading teams.

• Bachelors degree or equivalent experience.

• Results-focused leadership: you build the process, hit the dates, and hold the quality bar.

Ways to stand out from the crowd:

• Experience adopting Gaussian splats, NeRFs, neural reconstruction, neural shading, or other advanced rendering techniques for AV, robotics, simulation, synthetic data, or production rendering workflows.

• Hands-on experience building, adapting, or deeply evaluating world models, video generation systems, neural rendering systems, Gaussian splatting / NeRF pipelines, or neural shading techniques.

• Experience moving from earlier neural rendering approaches, such as NeRFs or Gaussian splatting, toward generative world models for interactive simulation, rendering, robotics, or AV.

• Published or peer-recognized work advancing the state of the art in world models, neural rendering, generative simulation, video generation, or production ML infrastructure.

• Experience building developer ecosystems or research communities around open-source projects.