Software DevOps Engineer, Networking

NVIDIA's networking software runs the fabrics inside the world's largest AI systems: NVLink and InfiniBand switch management, subnet managers, in-network computing, and fabric telemetry. Our infrastructure and DevOps team owns how that software is built, tested, scanned, and released — the CI/CD pipelines, Kubernetes-based build fleets, artifact and container registries, quality and security gates, and the hardware-simulation and lab environments the product teams rely on every day. We are looking for a senior DevOps engineer who is at home on Linux, automates in Python and Bash without hesitation, and takes pipelines end to end, from a red nightly to a signed release. The work spans a dozen active repositories and three CI systems, together with a distributed team of DevOps and product engineers across the US and Israel. Every build the networking software organization ships passes through what this role owns. What you'll be doing: • Design, build, and maintain CI/CD pipelines for a portfolio of networking software products across GitLab CI, Jenkins, and GitHub Actions — from merge-request gates to nightly regressions and release pipelines that package, sign, and publish Debian and RPM packages and container images. • Keep pipeline signal trustworthy: make failures visible and actionable, remove failure masking, stabilize flaky lanes, and turn nightly regression results into something product teams act on. • Operate and evolve the build infrastructure — Kubernetes-based CI runner fleets, container registries and artifact repositories, caching, retention policies, and access control. • Integrate quality and security gates into every pipeline: code coverage, static analysis, unit and functional test lanes, secret and container scanning, dependency and vulnerability scanning — and drive findings to closure with the owning teams. • Automate provisioning of hardware- and simulation-based test environments with Ansible and Python. • Partner with C++ and Python developers, verification engineers, and lab and infrastructure teams to unblock builds, harden release paths, and set repository standards — branch protection, code owners, merge policies. • Apply AI-assisted tooling to DevOps work — automated code review, failure-triage assistants, agent workflows — wherever it removes toil. What we need to see: • Bachelor's degree in Computer Science, Computer Engineering, or a related field, or equivalent experience. • 5+ years in DevOps, build and release, or infrastructure engineering roles, including pipelines owned end to end. • Deep, hands-on Linux skills on Ubuntu and RHEL: shell fluency, process and system debugging, package management, operating fleets of hosts, and containers with Docker. • Strong automation in Python and Bash. • Production track record with at least one modern CI system — GitLab CI, Jenkins pipelines, or GitHub Actions — including multi-stage pipelines for both compiled and interpreted codebases. • Working knowledge of Kubernetes as a CI and infrastructure platform, plus container registries and artifact repositories. • Familiarity with software quality and security tooling: coverage, static analysis, secret and vulnerability scanning. • Solid grounding in computer networking fundamentals and Git-based development workflows. • Ownership and clear communication — able to drive a problem across teams and time zones and to document what you build. Ways to stand out from the crowd: • Building and packaging C/C++ projects with CMake and Conan, and keeping large compiled builds fast and reproducible. • Familiarity with high-performance networking — InfiniBand, Ethernet fabrics, NVLink — or with switch and network-management software. • Running SonarQube, Coverity, Bullseye, or container and secret sca