Senior GenAI Research Engineer - Optimization and Kernels

Databricks

3 months ago

San Francisco, CA, USA

Senior

H1B Sponsor

Base Salary

$166k - $225k/yr

Responsibilities

Drive performance improvements through advanced optimization techniques.
Design, implement, and optimize high-performance GPU kernels for training workloads.
Create distributed training frameworks for large language models.
Profile, debug, and optimize end-to-end training workflows.

BS/MS/PhD in Computer Science or related field.
Hands-on experience writing and tuning CUDA kernels for ML training applications.
Strong understanding of NVIDIA GPU architecture and proficiency with CUDA debugging tools.
Deep understanding of parallelism techniques and memory optimization strategies.
Strong software engineering skills in Python and PyTorch.

PythonPyTorch