Staff Test & Release Engineer, Inference Cloud
Core
Design and build end-to-end test infrastructure and release processes for d-Matrix's Inference Cloud, a deployment and orchestration platform for generative AI models running on GPU and Corsair accelerator clusters.
Role type
Staff Test & Release Engineer (Inference Cloud)
Builds
Automated test suites, CI/CD pipelines, and release orchestration systems for inference services.
Domain
Generative AI, Cloud Infrastructure, GPU/Accelerator Orchestration
Deliverable
production ML models
Required skills
Test infrastructure design, CI/CD pipeline engineering, Kubernetes (deployment/debugging/Helm/operators), Python, pytest, shared hardware scheduling, release strategy definition, stakeholder consensus building, AI/LLM tooling fluency.
Preferred skills
LLM serving system testing (vLLM, SGLang, TensorRT-LLM), speculative decoding, hardware-in-the-loop testing, operator/controller testing tools (envtest, kuttl), canary rollouts, Rust.
Technologies
Kubernetes, Azure (AKS, Azure DevOps, GitHub Actions), GitLab CI, GitHub Actions, Buildkite, Argo, Python, Rust, Helm, vLLM, SGLang, TensorRT-LLM.
Responsibilities
Design end-to-end test infrastructure for homogeneous and heterogeneous inference deployments; convert user stories into automated end-to-end tests including failure cases; run tests on GPU/Corsair nodes in CI with fair scheduling; own release strategy for orchestration platform and Kubernetes tooling; lead component-level release strategy and compatibility testing; define release gates for functional, upgrade, accuracy, and performance tests; partner with QE and platform teams on shared test strategy.
Seniority
Staff, hands-on IC with strategic scope.
