Broadcom
AI Systems Performance Engineer
$144K–$230K
USA-CA San Jose Innovation Drive · Full-time · Posted Oct 9, 2026
- Experience
- Typically 12+ years of related industry experience; candidates with fewer years and an exceptional track record are encouraged to apply.
- Education
- Bachelor’s or Master’s degree in Computer Science, Computer Engineering, Electrical Engineering, or a related field.
- Visa sponsorship
- Must have authorization to work in the US.
About the job
Benchmarks and tunes NIC and host networking performance for AI workloads on GPU clusters, diagnosing bottlenecks and reporting findings to guide deployments and product positioning.
What you'll do
- Run AI benchmarks, including MLPerf and NCCL tests, and make benchmarking repeatable with automation.
- Tune NICs, drivers, and host networking for distributed AI workloads; diagnose performance bottlenecks across Linux, hardware, and networks.
- Document methodologies and present findings and recommendations to hardware, software, and networking teams.
What they ask for
- Bachelor’s or Master’s degree in a related field.
- Strong hands-on Linux experience, including system-level performance tuning and troubleshooting.
- Proficiency in Python or a similar scripting language.
- Deep experience in NIC/host networking, AI cluster building, or root-cause diagnosis of distributed-system performance problems.
Nice to have
- Experience with MLPerf Training or Inference; familiarity with Ethernet fabric behavior.
- Experience with Docker, Kubernetes, Slurm, or CI/CD pipelines for performance regression testing.
Pay
USD 143,800.00 To USD 230,000.00 (base salary)
Tools
Linux, Python, MLPerf, NCCL, RDMA/RoCEv2, Docker, Kubernetes, Slurm
AI focus
Benchmarks AI training, inference, and storage workloads on GPU clusters and tunes networking for distributed AI workloads.
Model-written summary · the original posting is authoritative