Infrastructure Engineer
WaveHouse is an open source API gateway for ClickHouse. WaveHouse Cloud is the hosted version, running on AWS with Terraform and EKS. You'd run that platform: provisioning, scaling, monitoring, deploys, backups, and the cloud bill. It's a DevOps and infrastructure job, but there's a fair amount of Go in it too, since the control plane is real software and not just config.
Responsibilities
- Run our Terraform and EKS setup
- Provision and upgrade per-tenant ClickHouse and WaveHouse instances
- Own backup and restore, and be able to prove it works before a customer needs it to
- Handle scaling when one tenant's load looks nothing like the next one's
- Build metrics, traces, logs and alerts that someone would actually act on at 3am
- Get deploys boring enough that we stop thinking about them
- Watch the cloud bill, especially egress
- Write the runbooks, so the next person on-call isn't guessing
- Share on-call with our CTO, and have a lot of say in what that rotation turns into
- Write Go on the control plane
- Own the infrastructure half of SOC 2
Qualifications
- You've run Kubernetes in production somewhere small enough that you were the one getting paged
- Real Terraform experience. Managing state, not following a tutorial
- Linux and AWS: EC2, VPC, IAM
- You've owned backup and recovery for something with real data in it
- You write Go, or you've written enough C, C++, Rust or Java that you'll be fine inside a month
- You've been on-call, and you can tell us about a night that went badly
- You're authorized to work in the US. We can hire on OPT and are open to supporting STEM OPT, but we can't sponsor or transfer H-1B visas right now
- Nice to have: a strong opinion about Flux vs Argo
- Nice to have: Prometheus and Grafana at a scale where cardinality became your problem
- Nice to have: secrets management. We're weighing Vault, OpenBao and the AWS native options and haven't committed to any of them
- Nice to have: Helm, operators, admission policies, Cilium, Karpenter
- Nice to have: you've run ClickHouse, or Druid, or Influx, or anything else columnar
- Nice to have: you've cut a cloud bill roughly in half and remember how
- Nice to have: a homelab
- Nice to have: open source infrastructure work
Links to open-source work, active GitHub profiles, or a homelab writeup are very welcome. No cover letter required.