Databricks
Staff Software Engineer - GenAI inference
$191K–$233K
San Francisco, California · Staff · Full-time · Posted Oct 8, 2025
- Experience
- 6+ years or equivalent in performance-critical systems.
- Education
- BS/MS/PhD in Computer Science or a related field.
- Visa sponsorship
- If access to export-controlled technology or source code is required, the employer may choose whether to apply for a U.S. government license and may decline to proceed with an applicant on that basis.
About the job
Leads the architecture, development, and optimization of Databricks’ GenAI inference engine and model-serving systems for large-scale LLMs.
What you'll do
- Design and build scalable inference capabilities, including routing, batching, scheduling, memory management, and distributed orchestration.
- Optimize latency, throughput, memory efficiency, and hardware utilization; develop profiling and tracing tools to identify bottlenecks.
- Work with researchers and cross-functional teams to productionize model advances and ensure reliable, fault-tolerant inference pipelines.
What they ask for
- 6+ years or equivalent in performance-critical systems.
- Strong knowledge of ML inference internals and hands-on experience with CUDA and GPU programming.
- Strong background in distributed systems design.
- Experience building instrumentation, tracing, and profiling tools for ML models.
Nice to have
- Published research or open-source contributions in ML systems, inference optimization, or model serving.
Pay
$190,900–$232,800 USD base salary
Tools
CUDA, cuBLAS, cuDNN, NCCL
AI focus
Builds and optimizes LLM inference systems, bringing new model architectures and features such as sparsity, activation compression, and mixture-of-experts into production.
Model-written summary · the original posting is authoritative