Infrastructure Engineer

Era is building the AI orchestration layer for the next generation of physical devices. We power intelligence across headphones, wearables, home objects, and entirely new device categories, and our infrastructure needs to be as ambitious as the vision.

We've raised $10M+ from Abstract, Topology, BoxGroup, Betaworks, Collab Fund, and Mozilla Ventures.

---

The Role

Era's platform is model-agnostic, framework-agnostic, and environment-agnostic by design. We orchestrate intelligence across many LLM providers, any device form factor, and any deployment topology our partners need. The infrastructure underneath all of that needs to be just as flexible and modular, as it does unbreakable.

You'll own the foundational infrastructure that everything at Era runs on. That means our multi-cluster Kubernetes platform, our real-time voice pipeline, our device telemetry ingestion, and our GitOps deployment machinery. But it also means something bigger: you'll be designing Era's infrastructure to be self-deploying, self-monitoring, and self-healing - an automated, agentic approach to infrastructure operations that is among the first in the world to operate at this scale.

What You'll Do

• Own the infrastructure underpinning the Era Core platform - multi-cluster Kubernetes across production, staging, and tools environments, with GitOps overlays, ingress routing, model hosting, and automated TLS
• Design self-healing, agentic infrastructure. Build the systems that let Era's infrastructure monitor itself, auto-remediate failures, scale proactively, and deploy without human intervention
• Operate and evolve our GitOps pipeline, from source to production with security scanning, rolling deployments, and environment promotion

•Scale the pod pool system - our pre-warmed Kubernetes pod pool provisions AI agent environments in <2>

• Build and operate device fleet infrastructure - telemetry ingestion, OTA update delivery, device provisioning, and health monitoring across a growing fleet of Era-powered hardware
• Run our real-time voice infrastructure - WebRTC servers with TURN gateways and autoscaling
• Build storage and data infrastructure - relational databases, time-series analytics, caching layers, and vector search
• Own observability and incident response - metrics, structured logging, health checks, and analytics dashboards
• Optimize cost relentlessly - Kubernetes node pools, LLM inference routing, pod resource limits, and container registry lifecycle

What We're Looking For

• Deep production Kubernetes experience. You've managed multi-cluster deployments, debugged pod scheduling, tuned resource requests/limits, and operated at meaningful scale
• Device fleet and edge infrastructure experience. You've built or operated systems ingesting telemetry, pushing OTA updates, or managing provisioning across 100k+ physical devices
• Strong understanding of GitOps patterns - ArgoCD or Flux, Helm or Kustomize, infrastructure-as-code
• Experience building infrastructure for real-time, low-latency systems - WebRTC, WebSocket, audio streaming, or similar
• Hands-on with PostgreSQL and vector databases (Weaviate, Milvus, etc) operations at scale - replication, connection pooling, migration strategies
• You think in cost curves. You've optimized cloud spend and know how to keep infrastructure costs from blowing out of proportion
• CI/CD pipeline design. You've built and maintained pipelines that teams actually trust and use
• Comfort with ambiguity. We're building infrastructure for a product category that doesn't have established playbooks

Nice to Have

• Experience with LLM inference infrastructure - model serving, response caching, multi-provider routing, token cost optimization
• Background in autonomous/self-healing infrastructure - chaos engineering, auto-remediation, AIOps
• Authentication/identity platform operations (Ory, Auth0, or equivalent)
• Experience building custom developer tooling - internal CLIs, deployment automation, developer experience infrastructure
• Experience at a company that scaled infrastructure from early stage to 100k+ users

You at Era

We're at the very beginning of something massive. You won't be maintaining someone else's infrastructure, you'll be designing it from first principles for a problem no one has solved before: how do you run real-time AI across millions of physical devices, across dozens of model providers, at a cost that actually works? You'll be among the first infrastructure engineers in the world to build truly agentic infrastructure - systems that don't just run, but operate themselves.

SF. We have teams in SF and Australia. Competitive salary, healthcare, and meaningful equity.