Senior Platform Engineer

NikoHealth•New York, NY

3d•$150,000 - $200,000•Remote

About The Position

This is a newly created role, and one of the first two engineering hires NikoHealth will make in the United States. You'll report directly to the CTO and own how our software gets released and run while working closely with other senior engineers. We operate dozens of .NET services deployed per-customer on Windows Server and IIS in AWS, managed with Terraform and Octopus Deploy. Across hundreds of provider tenants that's thousands of running application instances. Today a change waits for the next release train, and the train then takes weeks to reach the whole fleet, so end-to-end a given feature takes months. Improving our deployment and development experience is the first job. After that there are dozens of areas of platform investment needed to support how fast we're scaling.

Requirements

5+ years in platform, infrastructure, or DevOps engineering.
Production Terraform, including state you owned and drift you had to reconcile.
Production Windows Server and IIS, application pools included.
End-to-end ownership of a CI/CD pipeline in a previous role.
AWS at the level this estate uses it: EC2, ALB, VPC, IAM.
A track record of replacing a manual process with a measured and automated one, where you can tell us the before and after.
Evidence you've made other engineers faster, not just the systems. Build times, local setup, test feedback, whatever it was.
Daily working use of AI coding agents such as Cursor or Claude Code, including the harness around them: tests, project instructions, and a loop you can run again.

Nice To Haves

Octopus Deploy, or another deployment orchestrator you owned.
Single-tenant or multi-tenant SaaS at meaningful scale.
Healthcare, HIPAA, or another regulated environment.

Responsibilities

Own our CI/CD pipeline end to end, with the authority to change what it deploys and how.
Own developer experience. Reproducible builds, a local environment someone can actually stand up, and fast honest feedback on every change. This is half the job, not a side effect of the other half.
Get our automated regression suite running in CI. It exists, it's substantial, and it isn't currently executed.
Cut release lead time and get the fleet inside a published, committed release window.
Make fleet version state answerable by a query instead of by a person, so anyone can see what a customer is running.
Design and publish an access model, so provisioning stops routing through one inbox.
Start bringing runtimes and dependencies current, beginning with the shared libraries everything else depends on.
Document and share what you learn, so infrastructure knowledge stops being concentrated in one place.