Cerebras
Staff GPU Inference SDET
Sunnyvale, CA +1 · Hybrid · Staff · Full-time · Posted Sep 9, 2026
- Location
- Sunnyvale, CA
Toronto, CAN - Experience
- 8+ years of software engineering experience as an SDET, Infrastructure Quality Lead, or Systems Test Engineer.
About the job
Build and lead automated release qualification, testing, and reliability systems for multi-node GPU inference infrastructure and serving workloads.
What you'll do
- Develop automated test frameworks, regression gates, and release pipelines for the GPU inference stack.
- Benchmark serving workloads and validate performance, numerical correctness, and output quality.
- Create fault-injection, resilience, and observability systems for GPU clusters and integrate tests into CI/CD.
What they ask for
- 8+ years of software engineering experience as an SDET, Infrastructure Quality Lead, or Systems Test Engineer.
- Hands-on experience provisioning and validating multi-node GPU clusters in public cloud or enterprise data centers.
- Deep knowledge of LLM serving engines and distributed runtimes, including prefill/decode disaggregation, KV-cache management, and dynamic batching.
- Expert Python skills and experience building test automation frameworks, diagnostic tools, and CI/CD integrations.
Nice to have
- Direct experience with AMD (ROCm / HIP) or NVIDIA software stacks.
- Experience with workload replay, ML evaluation pipelines, MLPerf Inference, kernel profiling tools, or C++.
Tools
Python, Kubernetes, Slurm, Ray, InfiniBand, RoCE, NCCL, Prometheus
AI focus
Validate LLM inference serving, including workload performance, model accuracy, precision stability, determinism, and output correctness.
Model-written summary · the original posting is authoritative