Fundamental Generative AI Research (GenAIR) Group

Welcome to the homepage of NVIDIA’s Fundamental Generative AI Research (GenAIR) group, led by Arash Vahdat. We are part of the Learning and Perception Research (LPR) organization and mostly focus on generative learning and its applications in different areas. We are particularly interested in:

• Fundamental Generative AI: diffusion and flow models, efficient training and sampling, distillation and acceleration, semantic and modular control, test-time scaling, RL-based post-training, equivariant models, discrete and continuous generative models, new data modalities.
• Generative AI for Biology and Chemistry: protein and biomolecule design, binder design, molecular generation and optimization, reaction modeling, synthetic pathway planning, lab-in-the-loop learning, wet-lab alignment, test-time steering, agentic scientific discovery.
• Video and World Models: video generation, world models, accelerated generation, long-video generation, efficient memory, interactive and streaming generation.
• Diffusion Language Models: continuous and discrete diffusion language models, parallel decoding, inference acceleration, distillation, post-training, test-time scaling, reinforcement learning, reasoning, speculative decoding.

We are currently looking for outstanding candidates to join our team as:

• Research Interns for 2027: Apply Here.
• Research Scientist: general fundamental generative AI scientist or biomolecular design scientist: Link.
• Senior Research Scientist: Apply here if you finished your PhD more than one year ago: Link

For general inquiries, you may email your resume to genair-openings@nvidia.com. Kindly note that we may not be able to respond to all inquiries via email.