Reactor is building the inference platform for real-time world models. Our goal is to democratize access to real-time intelligence to unlock next-generation use cases in media, physical AI and omni-directional models. We already run the largest and fastest catalog of world models on the market, and work closely with all major model labs to run and optimize their world models, including Alibaba, Amazon, and NVIDIA. Our team is made of alumni from leading firms such as Apple, Meta, Amazon, Microsoft, Luma AI, Adobe, Stability AI, and more. We've raised $74M led by Lightspeed, with Amplify Partners, WNDR, NVIDIA, Sapphire, and other Tier 1 investors participating. Jeffrey Katzenberg, co-founder of DreamWorks Animation, sits on our board of directors.
We're looking for a senior cloud and platform engineer who builds the foundation everything at Reactor runs on, and owns keeping it up.
That foundation is big for our size. Production runs across clouds, in regions from São Paulo to Tokyo, on many types of GPUs. Every new model and every new customer lands on it.
This isn't a pure DevOps or SRE seat. You'll design systems and write real code, and you'll also own the deploys, the on-call and the cleanup. Our ML Infrastructure team builds the inference scheduling and serving layer on top of what you run.
We're in person in San Francisco.
We're looking for a senior cloud and platform engineer who builds the foundation everything at Reactor runs on, and owns keeping it up.
That foundation is big for our size. Production runs across clouds, in regions from São Paulo to Tokyo, on many types of GPUs. Every new model and every new customer lands on it.
This isn't a pure DevOps or SRE seat. You'll design systems and write real code, and you'll also own the deploys, the on-call and the cleanup. Our ML Infrastructure team builds the inference scheduling and serving layer on top of what you run.
We're in person in San Francisco.
What You'll Do
- New clouds and regions. Bringing up GPU providers and regions in Terraform, and making the next one a copy-paste instead of a project
- Kubernetes across all of it. Cluster lifecycle, autoscaling (HPA, VPA, node pools), capacity fallbacks when a region runs out of GPUs
- The real-time network. Ingress, load balancing, our WebRTC media relay, and geo-routing users to the closest GPU
- Data and messaging. Kafka and Redis running across regions, with alerting that catches problems before customers do
- Ship path. Flux GitOps, Helm charts, image automation, and CI/CD for a Go and Python monorepo
- Observability. Grafana, Prometheus, OpenTelemetry and eBPF tracing across services and GPU workloads
What We're Looking For
- 6+ years building and running production infrastructure, with real scale behind it
- You've operated Kubernetes in production, not just deployed to it: debugged scheduling, tuned autoscalers, run cluster upgrades
- Strong infrastructure as code (Terraform, Pulumi or similar) across many environments and regions
- Solid distributed systems fundamentals, and you write production code in Go or Python
- Networking: load balancers, TLS, DNS, NAT, cross-region latency
- You've owned infrastructure end to end at a startup
- Systems judgment (failure modes, blast radius, cost)
Nice to Have
- GPU workloads on Kubernetes: device plugins, GPU-aware scheduling, bin-packing expensive hardware
- GPU clouds beyond AWS
- Kafka or other streaming systems at scale
- Real-time media, WebRTC or low-latency networking
- GPU cost and capacity planning
Benefits
- Salary and equity at the 99th percentile for our stage
- Visa sponsorship and relocation support
- Generous health, dental, and vision coverage