Senior Software Engineer (AI)

TL;DR

Senior Software Engineer (AI Runtime): Building and scaling a managed GPU training platform for large-scale AI model training and fine-tuning with an accent on multi-node orchestration and distributed parallelism. Focus on optimizing training throughput, ensuring system resilience through failure detection, and maximizing GPU utilization across diverse hardware.

Location: Must be based in Mountain View or San Francisco, California

Salary: $160,000 — $225,000 USD

Company

Databricks is a data and AI company providing a Data Intelligence Platform used by over 10,000 organizations to unify data, analytics, and AI.

What you will do

• Drive the architecture and evolution of the AI Runtime (AIR) managed GPU training platform for fleets of thousands of accelerators.
• Solve complex challenges in multi-node orchestration, distributed parallelism strategies, and GPU scheduling.
• Optimize GPU efficiency, increasing model FLOPs utilization and overall end-to-end throughput.
• Develop resilience and observability foundations to detect and recover from hardware and software failures automatically.
• Collaborate with product and research teams to design the APIs, CLI, and developer experience for production training jobs.
• Mentor other engineers and lead end-to-end engineering efforts from design to production rollout.

Requirements

• 5+ years of experience building large-scale distributed systems, GPU training infrastructure, or ML systems.
• Proficiency with distributed training frameworks such as PyTorch, FSDP, DeepSpeed, or Megatron.
• Deep understanding of GPU performance, including accelerator architecture, NVLink, InfiniBand, or RoCE.
• Experience operating managed multi-tenant cloud platform products with strict SLAs and SLOs.
• Strong foundation in algorithms, data structures, and performance-sensitive system design.
• BS in Computer Science or a related field (MS or PhD preferred).

Culture & Benefits

• Comprehensive benefits and perks tailored to the region.
• Opportunity to work on frontier-scale foundation models and cutting-edge AI infrastructure.
• Collaborative environment partnering across product, research, and platform teams.
• Commitment to diversity, inclusion, and equal employment opportunity.