Senior Cloud & Platform Engineer — Real-Time Inference

Reactor; San Francisco (CA); On-site; USD 180,000 - 260,000; Full time

Benefits offered by this job

Health, dental, and vision coverage

Salary and equity at top percentile

Job summary

Reactor is building the inference platform for real-time world models and is seeking a senior cloud and platform engineer to own the foundation that everything runs on. Production spans multiple clouds and regions, supports GPUs, and serves new models and customers worldwide.

You’ll design systems, write production code in Go or Python, own deploys and on-call rotations, and contribute to the ML Infrastructure team's real-time scheduling and serving stack.

Qualifications

• 6+ years building and running production infrastructure with real scale.
• You've operated Kubernetes in production and know how to debug scheduling and autoscalers.
• Strong infrastructure as code across environments and regions (Terraform, Pulumi or similar).
• Production code in Go or Python.
• You've owned infrastructure end-to-end at a startup.
• Systems judgment: failure modes, blast radius, cost.

Responsibilities

• New clouds and regions: bring up GPU providers and regions in Terraform, duplicable templates.
• Kubernetes across the stack: cluster lifecycle, autoscaling, region GPU capacity fallbacks.
• The real-time network: ingress, load balancing, WebRTC, geo-routing to nearest GPU.
• Data and messaging: Kafka and Redis across regions with proactive alerting.
• Ship path: Flux GitOps, Helm charts, image automation, and CI/CD for a Go/Python monorepo.
• Observability: Grafana, Prometheus, OpenTelemetry and eBPF tracing across services and GPUs.

Skills

Kubernetes

Go

Python

Distributed systems

Infrastructure as code

Startup experience

Tools

Terraform

Pulumi

Grafana

Prometheus

OpenTelemetry

eBPF

Job description

Reactor is building the inference platform for real-time world models and is seeking a senior cloud and platform engineer to own the foundation that everything runs on. Production spans multiple clouds and regions, supports GPUs, and serves new models and customers worldwide.

You’ll design systems, write production code in Go or Python, own deploys and on-call rotations, and contribute to the ML Infrastructure team's real-time scheduling and serving stack.