As a Staff Software Engineer for AI Runtime at Databricks, you will be instrumental in building and scaling systems for large-scale GPU training. This role focuses on enhancing the architecture and performance of the managed GPU training platform, ensuring reliability and efficiency. You will collaborate with various teams to define the technical vision and mentor other engineers. Your contributions will directly impact the effectiveness of custom training solutions for AI teams worldwide.

• 10+ years of experience building and operating large-scale distributed systems, particularly in GPU training infrastructure or ML systems.
• Hands-on experience with distributed training frameworks such as PyTorch, FSDP, DeepSpeed, or Megatron.
• Strong understanding of training resilience patterns, including checkpointing and automatic recovery for long-running jobs.
• Solid grasp of GPU performance fundamentals, including accelerator architecture and high-speed interconnects.
• Experience building and operating managed, multi-tenant platform products in the cloud.
• Strong foundation in algorithms, data structures, and system design for performance-sensitive systems.
• Proven ability to deliver technically complex initiatives that create clear customer value.
• Strong communication skills and ability to collaborate across teams.
• Strategic mindset with a passion for mentoring engineers.
• BS in Computer Science or a related field; MS or PhD preferred.

• Drive the architecture and evolution of AIR's managed GPU training platform.
• Solve complex problems in large-scale training, including multi-node orchestration and GPU scheduling.
• Enhance GPU efficiency and training performance while lowering costs.
• Build resilience and observability foundations for multi-node jobs.
• Partner with product, research, and platform teams to improve the developer experience.
• Lead end-to-end engineering efforts from design to production rollout.
• Make high-impact contributions to core systems and support for new accelerators.
• Champion engineering excellence and mentor other engineers.