Top Tech Jobs & Startup Jobs in San Francisco Bay Area, CA

Lead GPU and NVLink-based cluster design and validation for large-scale AI and HPC deployments. Advise cloud partners on architectures, perform performance modeling, debug deployment issues, support NPI rollouts, and relay field feedback to engineering.

Design and implement functional GPU architecture models and simulation platforms. Develop tools, tests, and validation infrastructure, collaborate with architecture and engineering teams, and apply AI/ML to improve simulation and analysis workflows.

The Senior Performance Compiler Engineer will design and implement compiler technology for optimizing AI performance on NVIDIA GPUs, collaborating with teams to enhance efficiency in AI applications.

Lead hardware-software co-design for LLM inference: drive and document architecture, build executable C++ hardware models, optimize performance with good algorithms and parallelism, diagnose cross-team chip and subsystem issues, and automate workflows to improve efficiency.

Drive developer enablement for CUDA-Q by creating docs, examples, tooling, MCP servers, Agent Skills, prompt templates, and benchmarks. Analyze developer and agent workflows, define documentation and agent-consumable standards, run evaluations and metrics, and translate research into practical guidance to improve developer time-to-value for quantum and GPU-accelerated applications.

Develop datasets, models, and evaluation systems for content safety, product security, robustness, and ML fairness for multimodal LLMs. Research and implement bias detection/mitigation, define safety metrics, contribute MLOps automation and safety tooling, and collaborate with engineers, researchers, and data scientists to improve scalable model safety and inclusivity.

Lead a small team building a new Linux kernel-mode GPU driver: design architecture, implement device/kernel-mode code, collaborate with and upstream patches to the Linux kernel community, and represent NVIDIA in open-source discussions.

Lead research and engineering to develop, train, fine-tune, and deploy multimodal agentic foundation models. Advance post-training alignment (instruction tuning, RLHF/RLVR), develop agentic reasoning and grounded perception, and drive multimodal dataset collection, benchmarking, and evaluation for large-scale production models.

Design, develop, and maintain Linux graphics driver features for NVIDIA GPUs, focusing on Vulkan, OpenGL, and Wayland support. Collaborate with hardware, architecture, OEMs, and test automation to ensure performance, reliability, and timely regression detection across Linux platforms.

Develop and maintain GPU performance foundation libraries and test frameworks for Nsight tools. Implement GPU assembly tests, use emulators/simulators to validate instruction events, and document data workflows. Collaborate with hardware architecture teams to enable performance analysis and optimization of GPU-accelerated applications.