Thunder Compute VMWare for GPUs, monetized with a cloud platform. It's a defensible niche, post-PMF, with revenue growing 100%+ MoM. We have raised $4.5M to assemble a team of the world's top systems researchers and infrastructure engineers.

Company

Thunder Compute is the cheapest, easiest GPU cloud for developers. We’re a 4-person, seed-funded team (approaching Series A) with 100%+ MoM growth. 100% in-person, 6 days per week in SF. Our virtualization stack exposes network-attached GPUs over TCP, letting us oversubscribe hardware and pass savings to users.

Role

Own DevRel end-to-end. You’ll build and grow our community, ship hands-on demos and templates, teach developers how to run real workloads on Thunder, and bring sharp product feedback straight from Discord to the roadmap. High autonomy, high impact. You’ll report to the CEO.

What you’ll do

• Lead community: run Discord, office hours, meetups; turn user questions into guides and product fixes.
• Create content: quickstart templates, reproducible demos, blog posts, talks, and short videos.
• Be the SME for AI devs on Thunder: model hosting, training, inference, cost/perf trade-offs.
• Close the loop with product: file crisp bug reports, propose improvements, validate fixes.
• Prototype: small OSS tools, CLIs, or examples that unblock common workflows.
• Measure what matters: signups → activation, template adoption, community health.

You bring

• Clear writing and the ability to explain complex systems simply.
• Track record building developer communities and shipping educational content.
• Hands-on coding (Python preferred) for demos and scripts; comfortable with containers/cloud.
• Bias to action, resilience, and ownership from day one.
• Prior DevRel/advocacy experience at a startup or dev-tool company (show links: posts, talks, repos).

Nice to have

• GPU/AI infra familiarity (PyTorch, inference servers, RAG, Ray/FSDP, vectors).
• CS fundamentals and/or a related degree.
• Open-source contributions.

Compensation & setup

Full-time salary + meaningful equity. Starts as a 2-month contract, in-person optional, will full-time returning 100% in-person. You’ll work closely with the founding team and help define our DevRel function from zero.

How to apply

Send 2–3 links (writing/talks/demos), plus a short note on a community initiative you led and its outcome when you apply.

Our software network attaches GPUs at the CUDA layer. The core systems work requires deep expertise in systems, low-latency optimization, and C++.

Our infrastructure stack is written in GoLang and uses Kubernetes to manage instances.

This is a rare opportunity to approach deeply challenging technical problems in a high-stakes production environment on an early-stage team.