The Role
We're looking for a DevOps Engineer to join a fast-moving, AI-native engineering team.
This is not a traditional ops role.
A large share of our code is written and shipped by AI agents. The platform underneath them has to keep up.
You'll be expected to help our dev teams ship, debug, and monitor fast, using AI wherever it fits.
This is a ground-floor seat on a startup with serious growth ambitions.
If you want to build the platform that lets a team of humans and agents ship at machine speed, this is your role
What you'll do
Run CI/CD end to end: GitHub Actions, Bazel builds, and Argo CD deploys. Make them faster, greener, and safe for agents to trigger unattended
Build and maintain the infrastructure as code behind the platform on GCP with Terragrunt and OpenTofu: GKE, VPC networking, IAM, Secret Manager, Cloud SQL
Run the data platform: Cloud Composer (Airflow), dbt, and the BigQuery layer the business reports on. Keep pipelines on schedule, cheap, and trustworthy
Keep production observable: Datadog and Grafana, alerts people trust, error tracking that points at the real fault
Put AI to work on ops: extend the agent-driven alert triage, wire Claude Code into pipelines, automate the toil you'd otherwise do by hand
Work with backend and data engineers to make services and pipelines resilient, observable, and easy to deploy
Requirements
Minimum 4 years of hands-on DevOps/SRE experience
Kubernetes in production: you've deployed it, upgraded it, and debugged networking on it
GCP: GKE, Cloud SQL, BigQuery, Secret Manager, IAM, VPC networking
Infrastructure as code with Terragrunt and OpenTofu (Terraform experience counts)
CI/CD depth: good knowledge of GitHub Actions, and GitOps on Kubernetes with Argo CD, Flux, or equivalent
DataOps: Airflow, dbt, and a warehouse like BigQuery. You've operated pipelines in production, not only written them
Monitoring and alerting with Datadog, Grafana, or similar
Comfortable scripting and reading code in Go, Python, or TypeScript
A pulse on what's actually breaking in production right now
Someone who thrives with responsibility, moves fast, and wants to build something big
Nice to have
Bazel: you've managed the build of a big monorepo
Crossplane
Istio or another service mesh in production
Temporal or another durable workflow engine
Helm chart authoring and Kubernetes operators
Supply-chain security tooling: Trivy, gitleaks, SBOM
Advantages