NVIDIA Dev Monthly Meetup: San Francisco edition

​Join the NVIDIA developer community for an evening of learning, sharing, and connection.

​This NVIDIA Developer Meetup brings together engineers, builders, and AI practitioners from across the ecosystem to explore what’s new, what’s working, and what’s next in AI. You’ll hear the latest updates on CUDA, inference, and Nemotron, alongside talks from NVIDIA technical staff and members of the NVIDIA developer community sharing real-world insights and use cases.

​Agenda
6:00 pm — Doors Open
6:30 pm — Presentations from NVIDIA experts, partners, and developer community speakers including:

• ​

  • ​CUDA story time with Stephen Jones (Get to know Stephen here: https://www.youtube.com/watch?v=dNUMNifgExs)

  • ​The torch.compile integration in vLLM enables performance portability and clean separation between model implementations and lower-level optimizations across hardware platforms. In this talk, we'll explore the design of vLLM-compile and key fusion graph transformation passes, improving both runtime efficiency and developer productivity. We'll also preview in-progress work on reducing compilation time and a new LLM-specific compiler intermediate representation.

  • ​A brief presentation of SGLang’s Q1 roadmap, highlighting key priorities, upcoming initiatives, and milestones for the quarter.

  • ​Grove is a flexible Kubernetes API for deploying data-center scale workloads across training and inference. We'll dive into how Grove's features are used in Dynamo inference workloads on GB200 NVL72, as well as other production scenarios

​Whether you’re building, experimenting, or scaling AI solutions, this meetup is designed to spark ideas and conversation with peers tackling similar challenges.

​Space is limited—register early to save your spot.
Speaker lineup and full agenda coming soon!

​NVIDIA Privacy Policy: https://www.nvidia.com/en-us/about-nvidia/privacy-policy/