Kubernetes Platform Engineer, Infrastructure (GroqMetal) at Groq in New York, NY

Job description

Mission: GroqMetal is building the infrastructure platform that turns large-scale physical compute into a reliable, isolated, customer-consumable product. As a Kubernetes Platform Engineer, you will build and operate production software within the GroqMetal platform, helping deliver Kubernetes as a commercial product for customers running large-scale AI training and inference workloads. You will own meaningful platform components, work alongside experienced infrastructure engineers, and develop deep expertise in Kubernetes, distributed systems, and large-scale AI infrastructure. Location: This role will be based in one of our three hiring hubs: the Dallas, San Francisco, or New York City area . The person hired for this role must be based in one of these three areas. You’ll have the flexibility to work remotely while we establish our local Groq office, with the expectation that this role will transition to onsite once the office opens. Responsibilities & opportunities in this role: Build production Kubernetes platform capabilities: Implement APIs, controllers, operators, automation, and platform services that support the GroqMetal product by extending Kubernetes. Own components end to end: Take responsibility for well-defined platform capabilities from design and implementation through testing, deployment, operation, and continuous improvement. Operate what you build: Improve reliability, observability, performance, and operational efficiency, and participate in diagnosing and resolving production issues. Build a Kubernetes platform as a commercial platform: Develop capabilities used by dedicated customers running large-scale AI training and inference workloads. Develop distributed systems expertise: Build software that handles retries, failures, concurrency, reconciliation, and changing infrastructure state correctly. Improve the platform: Identify reliability, scalability, and usability problems and work with the team to deliver durable solutions. Collaboration is a core skill: Work within a team of experts to elevate the collective efforts of our team. Collaborate with engineers across infrastructure, inference, networking, infosec, and related teams to integrate platform capabilities holistically. Grow your technical scope: Contribute to design discussions, engineering practices, and technical decisions while developing toward ownership of increasingly complex platform systems. Ideal candidates have/are: Software engineers first: Experience designing, implementing, testing, shipping, and maintaining production software. Hands-on Kubernetes experience: Experience building or operating Kubernetes-based systems, including areas such as controllers/operators, CRDs, cluster lifecycle, node management, scheduling, networking, or platform services. Production infrastructure experience: Experience operating software in production and debugging problems across application, Kubernetes, networking, compute, and infrastructure layers. Solid distributed-systems fundamentals: An understanding of failure handling, retries, idempotency, consistency, reconciliation, availability, and recovery, with an interest in developing deeper expertise. Experience with disciplined software development: Comfortable with code review, automated testing, CI/CD, observability, deployment, and production operations. Product-minded infrastructure engineers: Experience building Kubernetes, cloud infrastructure, or platform capabilities consumed by customers is highly valued, but not required. Hands-on experience with modern development tools: Experience using coding agents such as Codex, Claude Code, or similar tools in substantive engineering work is valued. Interested in the AI infrastructure domain: Experience with hyperscalers, neoclouds, GPUs, accelerator infrastructure, or AI workloads is valuable, but candidates with strong platform engineering fundamentals can develop this expertise in the role. Comfortable growing through ambiguity: Able to independe