Senior Systems Software Engineer, Kubernetes Scale - DGX Cloud

NVIDIA's DGX Cloud organization delivers industry‑leading accelerated‑computing solutions for the most demanding AI workloads. The Senior Systems Software Engineer, Kubernetes Scale role sits at the heart of this effort, bringing together hardware expertise and open‑source software to make large‑scale AI infrastructure both performant and cost‑effective. In this position you will lead end‑to‑end performance and scalability studies of the DGX Cloud software stack, covering everything from the Kubernetes control and data planes to NVIDIA components such as the GPU Operator, Network Operator, DCGM, NIM, and distributed inference serving. You will work closely with AI researchers, developers and customers to create automated, realistic workload tests using custom and community tools, dive deep into complex performance issues, and design monitoring and analysis utilities that span GPU and CPU resources. Daily tasks also include triaging and debugging ultra‑large Kubernetes clusters, maintaining a high‑velocity CI/CD framework for continuous testing, documenting findings, and presenting results at internal meetings and external conferences like KubeCon and GTC. Engaging with upstream projects in the CNCF and NVIDIA open‑source ecosystems will be essential to validate and influence the scalability of AI workloads. The role requires at least eight years of experience with computer architecture, networking, storage systems or accelerators, together with a bachelor's or master's degree in electrical engineering, computer engineering, computer science or an equivalent background. Proven expertise in Kubernetes and related CNCF projects is mandatory, as is a solid track record working on large‑scale parallel and distributed accelerator‑based systems. Candidates must demonstrate the ability to optimise AI workloads at scale, conduct performance modelling and benchmarking, and write production‑grade code in Go or Python. Familiarity with the NVIDIA software stack for both training and inference, plus hands‑on experience with at least one public cloud provider (GCP, AWS, Azure, OCI, etc.), are also required. Preferred qualifications include strong operational experience with any Kubernetes distribution and prior success scaling clusters to ultra‑large node and object counts. A history of contributions to open‑source projects, excellent communication and interpersonal skills, and a PhD in a relevant discipline are considered advantageous. The position is based in Poland, where base salaries range from 292,500 PLN to 650,000 PLN depending on level and experience. NVIDIA is recognised as a top‑tier employer, offering a collaborative environment where creative, autonomous engineers can influence the future of AI infrastructure. What the role asks for: - 8+ years experience in computer architecture, networking, storage or accelerators - Bachelor's or Master's in EE, Computer Engineering, Computer Science or equivalent - Expertise in Kubernetes and related CNCF projects - Experience with large‑scale parallel and distributed accelerator systems - Proven ability to optimise AI workloads on massive systems - Experience in performance modelling and benchmarking at scale - Proficiency in Go or Python programming - Background with NVIDIA training and inference software ecosystem - Experience with at least one public cloud provider (GCP, AWS, Azure, OCI) - Strong operational experience with a Kubernetes distribution (nice‑to‑have) - Prior work scaling Kubernetes clusters to ultra‑large size (nice‑to‑have) - Demonstrated open‑source community contributions (nice‑to‑have) - Excellent communication and interpersonal abilities (nice‑to‑have) - PhD in a relevant field (nice‑to‑have)