Senior Systems Software Engineer, Accelerated Kubernetes Performance and Scale - DGX Cloud

Description

The company has spent more than 25 years building computer graphics, PC gaming and accelerated computing, and now applies its work to AI, where GPUs run computers, robots and self-driving cars.

The DGX Cloud group combines hardware and software to run accelerated computing for demanding AI workloads. The engineers there work on large, hard problems that shape how AI infrastructure runs in production.

The role focuses on scaling AI infrastructure while lowering total cost of ownership, reducing cost per token, and enabling future AI innovation and AI factories.

The company prefers hybrid work but will consider remote arrangements.

Responsibilities

• Lead end-to-end performance and scalability analysis of the Kubernetes-based accelerated runtime stack, covering both control and data planes and the company components such as GPU Operator, Network Operator, node-feature-discovery, topograph, dra-driver-nvidia-gpu and nvsentinel, tracing issues from orchestration down to the hardware.
• Design and contribute upstream architecture changes to the Kubernetes control plane and related projects so clusters run reliably at hyperscale, working in the open on problems hyperscalers usually solve privately.
• Reduce container startup and cold-start latency so inference scales smoothly across thousands of GPU nodes without putting pressure on the API server or adding operational fragility.
• Assess, improve and contribute to open-source projects that make Kubernetes a strong platform for AI workloads, such as Grove and gateway-api-inference-extension, designing their architectures for scalability, resilience and multi-node training and inference.
• Raise the scalability and performance of confidential containers (CoCo) on Kubernetes so encrypted inference meets strict efficiency and latency targets in production.
• Use DSX and related large-scale simulation infrastructure to model full AI-factory deployments and test scalability across thousands of simulated GPUs, finding failures that appear only at scale before the hardware arrives.
• Work with AI researchers, developers, customers and upstream communities to build automated at-scale workload tests, including replays of production agent traces, and monitoring and analysis tools, and add continuous performance and scale testing to modern CI/CD workflows.
• Document methods and results clearly, present findings internally and at industry events such as KubeCon and GTC, and engage with Kubernetes SIG Scalability, CNCF and the company open-source communities to shape and validate AI workload performance and scalability work.

Requirements

• A bachelor's or master's degree in engineering or equivalent experience, ideally in electrical engineering, computer engineering or computer science.
• 8+ years of experience in computer architecture, networking, storage systems and accelerator-based platforms.
• Expertise in Kubernetes and familiarity with the wider CNCF ecosystem.
• Deep experience with large-scale, parallel, distributed accelerator systems and with optimizing performance of AI workloads.
• Experience with performance modeling and benchmarking of large-scale systems.
• Proficiency in Golang and/or Python.
• Strong familiarity with the company software stack for training and inference.
• Expertise with at least one major public cloud provider, such as AWS, Azure, GCP or OCI.
• Strong operational experience with one Kubernetes distribution is a plus.
• Experience scaling Kubernetes clusters to ultra-large node and object counts is a plus.
• A record of working in open-source communities is a plus.
• Excellent communication and interpersonal skills are a plus.
• A PhD or equivalent experience in a relevant area is a plus.

Conditions

Hybrid work, with remote arrangements considered.

Base salary depends on location, experience and other objective job-related factors. Expected base salary is 184,000 USD to 287,500 USD for Level 4, and 224,000 USD to 356,500 USD for Level 5.

Applications are accepted at least until November 3, 2026.

This posting is for an existing vacancy.

The company uses AI tools in its recruiting process.

The company is an equal opportunity employer and does not discriminate on the basis of any characteristic protected by law.