Staff Software Engineer, AI Runtime: GPU Training Systems

Databricks Inc.; San Francisco (CA); On-site; USD 190,000 - 265,000; Full time

Job summary

Databricks is seeking a Staff Software Engineer to work on AI Runtime—the AI platform powering large-scale training across domains. You will drive architecture, reliability, and developer experience for multi-node GPU jobs, coordinating with research, product, and customers.

You’ll lead critical bets for future GPU generations, optimize performance, and mentor engineers to raise engineering excellence. This role is based in San Francisco with Databricks’ multi-office footprint.

Qualifications

• 10+ years in large-scale distributed systems.
• Hands-on experience training large models with FSDP, DeepSpeed, or Megatron.
• Fluency in parallelism strategies.
• Expertise in GPU performance, accelerators, interconnects and kernels.
• Background building and running multi-tenant cloud platforms as products.
• Track record leading technical projects across teams and delivering customer impact.
• Ability to align teams on technical designs and mentor engineers.
• BS in Computer Science or related field; MS or PhD preferred.

Responsibilities

• Drive major parts of AI Runtime’s architecture and help shape Databricks’ long-term technical direction in AI training infrastructure.
• Design reliability and observability systems for a growing, complex fleet.
• Lead improvements to AI Runtime’s developer experience for large training jobs.
• Help choose technical bets for new GPU generations and workload types.
• Tackle hardest multi-node training problems: orchestration, GPU scheduling, data loading, and checkpointing.
• Make training faster and cheaper by improving efficiency across models and GPUs.
• Collaborate with research, product, and customers to shape future releases.
• Mentor engineers and raise the standard of engineering excellence.

Skills

Distributed systems

GPU training

Parallelism strategies

GPU performance

Training resilience

Cloud platforms

Technical leadership

Mentoring

Communication

Education

BS in Computer Science or related field

MS or PhD preferred

Tools

FSDP

DeepSpeed

Megatron

Job description

Databricks is seeking a Staff Software Engineer to work on AI Runtime—the AI platform powering large-scale training across domains. You will drive architecture, reliability, and developer experience for multi-node GPU jobs, coordinating with research, product, and customers.

You’ll lead critical bets for future GPU generations, optimize performance, and mentor engineers to raise engineering excellence. This role is based in San Francisco with Databricks’ multi-office footprint.