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

Build end-to-end image generation and editing experiences inside ChatGPT. Implement interactive frontends, scalable backend services and APIs, prototype multimodal capabilities with researchers, optimize performance and reliability, and drive product launches through experimentation and iteration.

The Operations Program Manager leads hardware introductions and production ramps, ensuring timely execution, quality, and cost-effectiveness while coordinating cross-functional teams.

Design and implement advanced inference infrastructure for multimodal models, optimizing for performance and collaborating with researchers and product teams.

The Data Scientist, Safety will analyze safety-related data, improve classifiers, and design experiments to enhance OpenAI's product safety, partnering with cross-functional teams.

Lead design and implementation of a unified dataset read platform for large-scale training. Define dataset APIs, versioning, caching, reliability, and observability. Write and review core data-loading, service, and caching code, build visualizers and debugging tools, and coordinate across training frameworks, storage, and infrastructure teams.

Lead a team developing software systems to detect and prevent misuse of frontier AI models, ensuring production reliability and security.

Manage material readiness and supply chain execution for hardware programs, overseeing planning and coordination with cross-functional teams to ensure supply continuity and operational readiness.

Lead a small team to build performance modeling frameworks for AI systems, analyze architectural tradeoffs, and guide design decisions across various system components.

The 3P Architect will define AI infrastructure architectures, translate workload requirements, work with vendors, and lead proof-of-concept efforts while collaborating across teams and external partners.

Lead deployment and operations for OpenAI's custom silicon and systems in data center environments, ensuring successful deployment and ongoing reliability at scale.