Infrastructure Platform Engineer

About Massed Compute

Massed Compute is building a modern GPU cloud platform for AI and high-performance compute workloads. Customers choose us when they need high-performance infrastructure with more flexibility, visibility, and support than they get from traditional options. Our work sits at the intersection of AI infrastructure, product design, systems thinking, and customer obsession.

About the role

We run a multi-tenant GPU cloud: physical servers in several data centers, a virtualization layer on top of them, and the APIs customers use to rent capacity. This role covers both halves — you'll work in a hypervisor and a switch config one day and in a backend service and a SQL query the next.

Work is ticket-driven and AI-assisted. We expect engineers to use agentic coding tools as a normal part of the job, and to write clearly enough that the next person can follow the reasoning. Customer-reported problems land here too, so some of the work is direct: reproduce it, fix it, tell the customer what happened.

What you'll do

Run and maintain Proxmox clusters across multiple sites — nodes, networking, hardware faults

Build and maintain APIs and MCPs, including the public customer-facing surface

Keep the control plane and the actual infrastructure in agreement; automate away recurring issues

Investigate and resolve customer-reported issues, and communicate the resolution back

Fix multi-tenancy, security, and billing correctness issues as they surface

Work tickets end to end: diagnose, fix, verify, write it up

Evaluate vendors and architecture options, and document the decision

What you need

API development and maintenance

SQL — querying, schema work, and reasoning about transactions and concurrency

Strong Linux and networking fundamentals

Backend development in at least one server-side language

Hands-on experience using agentic coding tools — Claude Code, Cursor, Codex, or similar — as part of your normal workflow

Practical experience using MCP servers, and a working understanding of how they fit into an agentic workflow

Customer-facing troubleshooting — you can work a problem with the person who reported it and explain the outcome without jargon

Comfort with a ticket-driven workflow and clear written communication

3+ years in infrastructure, platform, or SRE-adjacent engineering

Nice to have

Proxmox (or comparable virtualization) administration, including its API

Building MCP servers

Experience in Node.js/JavaScript and Python

GPU infrastructure: NVIDIA drivers, device passthrough, hugepages, firmware

Data center hands-on work: cabling, switch configuration, procurement

Billing or metering systems

Object storage and FUSE performance work

OAuth 2.0, OpenTelemetry

On-call and incident command experience

How we'll know you're succeeding

Recurring alert classes shrink instead of holding steady

Infrastructure changes land cleanly, without customer-visible impact

Provisioning, teardown, and billing paths tell the truth about what they did

Fewer surprises reach customers before they reach us

Equal Opportunity

Massed Compute is an equal opportunity employer. We consider all qualified applicants without regard to race, color, religion, sex, sexual orientation, gender identity, national origin, age, disability, genetic information, veteran status, or any other characteristic protected by law.