Senior GenAI Research Engineer - Optimization and Kernels

About the role

The Databricks AI Research organization enables companies to develop AI models and agents using their own data, with technologies ranging from post-training open source LLMs to developing advanced multi-agent architectures. Databricks AI is committed to the belief that a company’s AI models and agents are just as valuable as any other core IP, and that high-quality AI should be available to all.

Responsibilities

• Drive performance improvements through advanced optimization techniques including kernel fusion, mixed precision, memory layout optimization, tiling strategies, and tensorization for training-specific patterns
• Create and optimize high-performance GPU kernels for training workloads (e.g., attention mechanisms, custom layers, gradient computation, activation functions) targeting NVIDIA architectures
• Design and implement distributed training frameworks for large language models, including parallelism strategies (data, tensor, pipeline, ZeRO-based) and optimized communication patterns for gradient synchronization and collective operations
• Profile, debug, and optimize end-to-end training workflows to identify and resolve performance bottlenecks, applying memory optimization techniques like activation checkpointing, gradient sharding, and mixed precision training

Requirements

• BS/MS/PhD in Computer Science or related field with hands-on experience writing and tuning CUDA kernels for ML training applications, or hands-on experience in distributed training frameworks (PyTorch DDP, DeepSpeed, Megatron-LM, FSDP)
• Strong understanding of NVIDIA GPU architecture (memory hierarchy, tensor cores, warp scheduling, SM occupancy) and proficiency with CUDA debugging/profiling tools (Nsight, NVProf)
• Deep understanding of parallelism techniques and memory optimization strategies for large-scale model training, with proven ability to debug and optimize distributed workloads
• Strong software engineering skills in Python and PyTorch, with experience supporting production training workflows and knowledge of LLM training dynamics including hyperparameter tuning and optimization strategies

Qualifications

• BS/MS/PhD in Computer Science or related field with hands-on experience writing and tuning CUDA kernels for ML training applications, or hands-on experience in distributed training frameworks (PyTorch DDP, DeepSpeed, Megatron-LM, FSDP)
• Strong understanding of NVIDIA GPU architecture (memory hierarchy, tensor cores, warp scheduling, SM occupancy) and proficiency with CUDA debugging/profiling tools (Nsight, NVProf)
• Deep understanding of parallelism techniques and memory optimization strategies for large-scale model training, with proven ability to debug and optimize distributed workloads
• Strong software engineering skills in Python and PyTorch, with experience supporting production training workflows and knowledge of LLM training dynamics including hyperparameter tuning and optimization strategies

Skills

• Hands-on experience writing and tuning CUDA kernels for ML training applications
• Experience in distributed training frameworks (PyTorch DDP, DeepSpeed, Megatron-LM, FSDP)
• Understanding of NVIDIA GPU architecture (memory hierarchy, tensor cores, warp scheduling, SM occupancy)
• Proficiency with CUDA debugging/profiling tools (Nsight, NVProf)
• Deep understanding of parallelism techniques and memory optimization strategies for large-scale model training
• Experience supporting production training workflows
• Knowledge of LLM training dynamics including hyperparameter tuning and optimization strategies

Benefits

Details on specific benefits are available upon request.

Pay

The pay range(s) for this role is listed below and represents the expected salary range for non-commissionable roles or on-target earnings for commissionable roles. Actual compensation packages are based on several factors that are unique to each candidate, including but not limited to job-related skills, depth of experience, relevant certifications and training, and specific work location. Based on the factors above, Databricks anticipates utilizing the full width of the range. The total compensation package for this position may also include eligibility for annual performance bonus, equity, and the benefits listed above.

Schedule

Details on specific schedules are available upon request.

Contact

If you have questions about this role, please contact [Contact Information].