Staff Software Engineer AI Runtime
Databricks is a data and AI platform that lets organizations build analytics, AI agents, and applications on a unified, governed lakehouse.
Funding history
Investors
About Databricks
Data engineers, analysts, and AI teams use Databricks to process large datasets, build reliable pipelines, and train models on a single governed platform. Users can run SQL analytics, serve ML predictions in real time, and deploy AI agents grounded in enterprise data. Its open lakehouse architecture provides consistent security and governance across analytical and operational workloads.
Skills
Candidate Availability
Required and preferred rules are kept separate and reflect the wording in the original posting.
About the Role
You will drive the architecture of a managed GPU training platform. You will improve distributed training performance, scheduling, resilience, observability, and developer workflows for launching and operating large-scale jobs. You will lead engineering initiatives from design through rollout, mentor engineers, and help shape APIs and tooling.
Requirements
- 10+ years of experience building and operating large-scale distributed systems
- GPU training infrastructure, high-performance computing, or ML systems experience
- PyTorch
- FSDP
- DeepSpeed
- Megatron
- Distributed training
- Checkpointing
- Failure detection
- Automatic recovery
- GPU architecture
- NVLink
- InfiniBand or RoCE
- Collective communication
- Cloud platform experience
- Algorithms
- Data structures
- System design
- Bachelor's degree in Computer Science or a related field
Responsibilities
- Drive the architecture and evolution of the managed GPU training platform
- Solve multi-node orchestration, distributed parallelism, GPU scheduling, data loading, and checkpointing challenges
- Improve GPU utilization, training throughput, and training cost efficiency
- Build resilience and observability for multi-node jobs
- Shape APIs, CLI tools, and developer workflows for production training jobs
- Lead engineering efforts from design through production rollout
- Support new accelerators and regions
- Mentor engineers through design reviews and technical discussions
Benefits
- Annual performance bonus
- Equity