Databricks

Staff Software Engineer - GenAI Performance and Kernel

Databricks

Apply
5 months ago
San Francisco, CA, USA
Staff+
H1B Sponsor

Base Salary

$191k - $233k/yr

Responsibilities

  • Lead the design and implementation of core compute kernels optimized for various hardware backends.
  • Drive the performance roadmap for kernel-level improvements including vectorization and auto-tuning.
  • Integrate kernel optimizations with higher-level ML systems.
  • Build and maintain profiling and verification tools to detect performance regressions.
  • Conduct performance investigations and root-cause analysis on inference bottlenecks.
  • Establish coding patterns and frameworks for kernel modularization and maintainability.
  • Influence system architecture decisions for effective kernel improvements.
  • Mentor and guide engineers in performance best practices.
  • Collaborate with teams to roll out optimizations into production and monitor their impact.

Requirements

  • BS/MS/PhD in Computer Science or a related field.
  • Deep experience writing and tuning compute kernels for ML workloads.
  • Strong knowledge of GPU/accelerator architecture and memory hierarchy.
  • Experience with advanced optimization techniques like tiling and vectorization.
  • Familiarity with ML-specific kernel libraries and open kernels.
  • Strong debugging and profiling skills using tools like Nsight and NVProf.
  • Experience with numerical stability, mixed precision, and quantization.
  • Experience integrating optimized kernels into real-world ML inference systems.
  • Proven track record of shipping performance-critical software.
  • Excellent communication and leadership skills.

Categories

AI & MLData Engineering