Research Engineer, Interactive World Models

Details

FlashDreams and FastGen are NVIDIA’s core technologies for turning video models into real-time world simulations. The stack spans model adaptation for faster generation and richer control, plus the execution layer that runs those models as responsive experiences. The technology supports autonomous-driving simulation, robot policy development and testing, game worlds, medical simulation, and virtual training.

As a Research Engineer, you will contribute across model development and runtime systems, building capabilities and helping turn research into systems that work in real applications. The work spans the world-model ecosystem, from emerging startups to established model labs.

Responsibilities

• Build and optimize the continuous autoregressive serving loop, including per-step control inputs, model and KV-cache state management, GPU inference, frame streaming, and model integrations.
• Advance production-ready world models by working with researchers on few-step distillation, causal or autoregressive generation, reward fine-tuning, action conditioning, and long-horizon spatiotemporal memory and consistency.
• Deliver multi-user experiences and simulation workflows from prototype through evaluation, integration, and release.
• Partner with applied researchers and domain teams to meet quality, performance, and reliability goals.

Requirements

• Experience in one or more areas such as video or world models, diffusion and generative modeling, model distillation and adaptation, simulation, robotics, computer vision, or real-time stateful machine-learning systems.
• A BS or MS in Computer Science, Computer Engineering, Electrical Engineering, or a related field, or equivalent experience.
• At least 3 years of relevant experience building, evaluating, integrating, optimizing, or serving machine-learning systems through industry, academic research, open-source work, or substantial projects.
• Strong Python and PyTorch skills, supported by software-engineering fundamentals in design, testing, debugging, version control, performance analysis, and Linux development.
• Ability to turn open-ended technical problems into working implementations, measure quality and performance, and communicate results clearly.

Preferred Qualifications

• Experience with post-training generative video models, including distillation, self-forcing, action conditioning, or long-horizon memory.
• Experience building and optimizing real-time, stateful generative inference systems, including history and KV-cache management, GPU kernels, quantization, parallel execution, streaming, scheduling, or multi-user serving.
• Contributions to an open-source machine-learning project or developer platform, such as implementing model support, improving performance, building tests and benchmarks, fixing difficult issues, writing documentation, or helping users adopt the technology.

Compensation And Benefits

The base salary range is $152,000–$241,500 USD for Level 3 and $184,000–$287,500 USD for Level 4. Base salary is determined based on location, experience, and the pay of employees in similar positions. Employees are also eligible for equity and benefits.

Applications will be accepted at least until August 24, 2026. NVIDIA uses AI tools in its recruiting processes. NVIDIA is an equal opportunity employer committed to an inclusive work environment.