Senior Software Engineer, AI Runtime
Senior Software Engineer building and scaling Databricks' managed GPU training platform (AI Runtime) for large-scale distributed AI model training. Requires 5+ years in distributed systems and hands-on experience with GPU training frameworks.
About the job
Responsibilities
- Drive the architecture and evolution of AIR's managed GPU training platform, delivering scalable, high-throughput, and resilient training across fleets that span thousands of accelerators.
- Solve the hardest problems in large-scale training, including multi-node orchestration, distributed parallelism strategies, GPU scheduling and dynamic routing, high-throughput data loading, and checkpoint and restore for very long-running jobs.
- Push GPU efficiency and training performance, raising utilization (such as model FLOPs utilization and end-to-end throughput) and lowering cost per training run across diverse model architectures and hardware generations.
- Build the resilience and observability foundations that keep multi-node jobs healthy, detecting and recovering from hardware and software failures with minimal disruption to customers.
- Partner with product, research, and platform teams to shape the APIs, CLI, and developer experience that make it easy to launch, monitor, and debug production training jobs.
- Lead end-to-end engineering efforts, from design through production rollout, holding a high bar for performance, correctness, and reliability.
- Make direct, high-impact contributions to the core systems behind AIR, and help bring up support for the latest accelerators and new regions as the fleet grows.
- Champion engineering excellence, mentor other engineers through design reviews and technical discussions, and contribute to Databricks' technical direction in AI training infrastructure.
Requirements
- 5+ years of experience building and operating large-scale distributed systems, with experience in GPU training infrastructure, high-performance computing, or ML systems.
- Experience with distributed training frameworks (such as PyTorch, FSDP, DeepSpeed, or Megatron) and the parallelism strategies (data, tensor, pipeline, and sequence parallelism) used to train large models.
- Strong understanding of training resilience patterns, including checkpointing, failure detection, and automatic recovery for long-running, multi-node jobs.
- Solid grasp of GPU performance fundamentals, including accelerator architecture, high-speed interconnects (such as NVLink and InfiniBand or RoCE), collective communication, and the bottlenecks that govern training throughput and utilization.
- Experience building and operating managed, multi-tenant platform products in the cloud, with clear SLAs and SLOs for availability, performance, and reliability.
- Strong foundation in algorithms, data structures, and system design as applied to performance-sensitive, large-scale distributed systems.
- Proven ability to deliver technically complex, high-impact initiatives that create clear customer or business value.
- Strong communication skills and the ability to collaborate across product, research, and infrastructure teams in a fast-moving environment.
- Customer-focused mindset with the ability to align implementation details with product goals, and a passion for mentoring engineers and fostering technical excellence.
- BS in Computer Science or a related field (MS or PhD preferred).
Nice-to-Haves
- MS or PhD in Computer Science or a related field.
Skills
PyTorch, Fsdp, Deepspeed, Megatron, Gpu Scheduling, Distributed Training, Checkpointing, Nvlink, InfiniBand, Roce, Collective Communication, Multi-Node Orchestration, High-Performance Computing, Ml Systems, System Design
Similar jobs
ML Engineering jobs
LangChain
New York, NY
Leads hands-on development and deployment of production AI agents and workflows, sets technical standards, and mentors engineers. Requires 4+ years of software engineering experience, production LLM application experience, and strong Python or TypeScript skills.
LegitScript
United States
Owns the full lifecycle of data and ML solutions, from ingestion and feature-ready datasets through production deployment and business-impact measurement. The role combines data engineering, applied machine learning, MLOps, and generative AI to build risk detection capabilities.
Mercury
San Francisco, CABuild and operate the platform that deploys, serves, observes, and retrains production machine-learning models for real-time fraud and financial-crime risk decisions. Requires 5+ years of ML engineering, backend, or MLOps experience, strong Python skills, and production model-serving expertise.
Parallel Systems
Los Angeles, CA
Develop and deploy real-time perception and sensor-fusion software for autonomous battery-electric rail vehicles. The role requires strong robotics, geometry-based computer vision, C/C++ and Rust experience, plus hands-on work with multimodal sensors and production systems.
Lightning AI
New York, NYBuild and optimize large language model training and post-training pipelines, improving model quality, distributed performance, evaluation, and production readiness. The role requires deep PyTorch and transformer experience, strong distributed-systems and software-engineering skills, and expertise in modern LLM optimization techniques.