Staff Kubernetes Platform Engineer, Infrastructure (GroqMetal)

Groq•San Francisco, NY

About The Position

Groq is building GroqMetal, an infrastructure platform that converts large-scale physical compute into reliable, isolated, customer-consumable AI infrastructure. This hands-on Senior Staff role will write production software as part of a software engineering team and lead the design of a Kubernetes-based platform for large-scale AI training and inference workloads. The role is a key technical leadership position in building systems that turn advanced physical infrastructure into a stable product. The role is based in the Dallas, San Francisco, or New York City area. Remote work is available while Groq establishes a local office, with the expectation that the role will transition to onsite work once the office opens. The position may require eligibility under applicable U.S. export control rules. Groq’s total cash salary ranges by level are $270,400–$365,000 for Staff and $341,400–$401,600 for Senior Staff. Placement depends on location, experience, skills, and internal compensation standards. International compensation varies by local market. Groq also offers a Long-Term Incentive Program and employee benefits.

Requirements

Strong experience designing, implementing, testing, shipping, and maintaining production software.
Deep Kubernetes platform engineering experience, including operators/controllers, CRDs, admission control, node lifecycle, cluster services, and production-grade platform design.
Practical distributed-systems expertise, including reasoning about CAP, consensus, leader election, eventual consistency, state reconciliation, fault domains, and recovery.
Experience developing infrastructure software through a robust SDLC, including design review, automated testing, CI/CD, staged rollout, rollback, observability, and incident learning.

Nice To Haves

Practical experience delivering Kubernetes, cloud infrastructure, or infrastructure control planes as a commercially critical or externally consumed product; valued but not required.
Substantive hands-on experience using coding agents such as Codex, Claude Code, or similar systems or harnesses.
Experience at a hyperscaler or neocloud supporting AI workloads or commercializing GPU clusters, with understanding of operational and product challenges of large-scale accelerator infrastructure.
Experience with inference platforms, accelerator/GPU infrastructure, device plugins, cloud infrastructure, or customer-facing platform products.

Responsibilities

Design and build core Groq platform systems, including APIs, controllers, operators, reconciliation loops, and infrastructure abstractions.
Write, review, test, debug, and operate substantial production code while providing Senior Staff technical leadership.
Use Kubernetes to build a commercial AI infrastructure product, delivering production-grade Kubernetes services for dedicated customers running large-scale AI training and inference workloads.
Apply distributed-systems principles to partial failure, retries, concurrency, consistency, idempotency, availability, and recovery.
Design stable platform APIs and contracts that decouple the customer experience from changes in backend infrastructure.
Collaborate across infrastructure, inference, networking, infosec, and other teams to turn ambiguous platform needs into production systems.