Senior GPU Platform Engineer

Description

About Zesty

Cloud teams should not have to choose between cost efficiency, performance, and control. At Zesty, we build automation that helps DevOps and FinOps teams optimize Kubernetes and cloud commitments continuously, within the guardrails they define.

We are a curious, practical team that likes solving hard problems together. You will have room to take ownership, share ideas openly, and see your work make a real difference for customers.

Position Overview

GPU workloads introduce complex challenges around utilization, allocation, reliability, and cost. We are looking for a Senior GPU Platform Engineer with hands-on DevOps/Platform and GPU optimization experience to build product capabilities that solve these challenges. You will join an R&D team and turn real-world infrastructure insight into production software.

Requirements

What you'll own

• Design and deliver GPU optimization capabilities, including improved GPU utilization, intelligent resource allocation, workload scheduling, and cost efficiency.
• Take ownership of complex features end-to-end - from technical design through production delivery.
• Translate real-world GPU infrastructure challenges into practical software solutions.
• Collaborate with R&D and product stakeholders to deliver reliable, high-impact product capabilities.
• Write clean, maintainable production code, leveraging AI-powered development tools to accelerate delivery, improve code quality, and streamline code reviews.

What will help you succeed

• More than 5 years of hands-on DevOps or Platform Engineering experience.
• Proven production experience optimizing GPU infrastructure or GPU workloads.
• Strong Kubernetes experience and demonstrated ability to write production-quality code.
• Evidence of measurable improvements in GPU utilization, performance, scheduling, allocation, capacity, reliability or cost efficiency.

It’s a plus if

• Experience with NVIDIA GPU Operator, DCGM, MIG, GPU sharing or time-slicing.
• Experience developing Kubernetes operators, controllers or scheduling-related components.
• Experience with GPU FinOps practices, including cost visibility, chargeback/showback, right-sizing or capacity planning.

Why this role matters

This is an opportunity to move from operating GPU infrastructure to building the product capabilities that solve those problems. The role has direct influence on how Zesty helps teams run GPU workloads more efficiently, reliably and cost-effectively. You will work on challenging cloud-native problems within an R&D team that values ownership.

A note on how we work -

We move fast, take ownership, and solve hard problems together. We value direct feedback, different perspectives, and the humility to keep learning as one team.