CoreWeave
Senior Software Engineer - Perf and Benchmarking
$182K–$242K
Sunnyvale, CA / Bellevue, WA +2 · Senior · Full-time · Posted May 14, 2026
- Location
- Sunnyvale, CA / Bellevue, WA
Bellevue, WA
Sunnyvale, California, United States - Experience
- 3–5 years.
- Visa sponsorship
- Applicants must be a U.S. person, eligible to access export-controlled information without export authorization, or eligible and reasonably likely to obtain the required export authorization; CoreWeave may decline to pursue export licensing for legitimate business reasons.
About the job
Build and operate Kubernetes-native benchmarking services and performance data systems, and support end-to-end MLPerf benchmarking across CoreWeave’s infrastructure.
What you'll do
- Develop services measuring latency, throughput, jitter, and cost-per-request; ingest, store, transform, and analyze performance events.
- Implement and maintain MLPerf Training and Inference workflows, including workload setup, cluster configuration, and result validation.
- Lead designs, improve service performance and reliability, and collaborate across teams while mentoring engineers and reviewing code.
What they ask for
- Strong programming skills in Python or Go; understanding of networked systems and performance fundamentals.
- Hands-on production Kubernetes experience and familiarity with CI/CD and observability tools.
- Experience with distributed systems, high-performance computing components, or cloud services.
- Effective cross-functional communication skills.
Nice to have
- Experience with time-series databases, LSM-based storage engines, or custom data pipelines; familiarity with MLPerf or other large-scale benchmarking frameworks.
- Contributions to OSS projects such as llm-d, vLLM or PyTorch; exposure to GPU-cluster benchmarking or GPU systems.
Pay
$182,000 to $242,000 USD base salary
Tools
Python, Go, C++, Kubernetes, Prometheus, Grafana, OpenTelemetry, MLPerf
AI focus
Benchmark AI/ML workloads and performance, including MLPerf Training and Inference runs and GPU systems or model-serving stacks.
Model-written summary · the original posting is authoritative