Staff Software Engineer (AI)

TL;DR

Staff Software Engineer (AI): Responsible for designing, implementing, and optimizing high-performance GPU kernels to power GenAI inference stack, leading the development of low-level compute paths and managing trade-offs between hardware efficiency and generality. Focus on kernel-level improvements, performance investigations, and root-cause analysis on inference bottlenecks to push the state-of-the-art in inference performance at scale.

Location: United States

Salary: $190,900—$232,800 USD

Company

Databricks is the data and AI company that unifies and democratizes data, analytics, and AI.

What you will do

• Lead the design, implementation, benchmarking, and maintenance of core compute kernels optimized for various hardware backends.
• Drive the performance roadmap for kernel-level improvements and integrate kernel optimizations with higher-level ML systems.
• Build and maintain profiling, instrumentation, and verification tooling to detect correctness and performance regressions.
• Lead performance investigations and root-cause analysis on inference bottlenecks.
• Establish coding patterns, abstractions, and frameworks to modularize kernels for reuse and maintainability.
• Collaborate with infrastructure, tooling, and ML teams to roll out kernel-level optimizations into production and monitor their impact.

Requirements

• BS/MS/PhD in Computer Science or a related field.
• Deep hands-on experience writing and tuning compute kernels (CUDA, Triton, OpenCL, LLVM IR, assembly or similar sort) for ML workloads.
• Strong knowledge of GPU/accelerator architecture.
• Experience with advanced optimization techniques.
• Strong debugging and profiling skills.
• Excellent communication and leadership skills.

Nice to have

• Published in systems/ML performance venues.
• Experience with custom accelerators or FPGA.
• Experience with sparsity or model compression techniques.

Culture & Benefits

• Committed to fostering a diverse and inclusive culture where everyone can excel.
• Hiring practices are inclusive and meet equal employment opportunity standards.
• Employees are considered without regard to protected characteristics.