About Us
Notion State is a premier Notion consultancy serving SMB and enterprise clients. Our portfolio includes Affirm, Typeform, HoneyBook, Articulate, Notion themselves, and many others. Being Notion experts is table stakes. We serve as embedded operational partners, spanning process strategy and design, implementation, and change management. Our aim is to be our clients' full-scope experts for organizing how they operate.
The Role
We're hiring a Backend Engineer, Integrations & Automations (fully remote, contract, ~20 hrs/week) to build the automations that extend what our clients' Notion workspaces can do, and to grow into owning that practice with us.
This is a build-first role with a clear runway. You'll start behind the scenes, partnering with our consultants who bring you client requirements. As you earn trust, the role expands: more hours, direct client ownership, and a leadership stake in our Worker practice as demand scales. We're not looking for a pair of hands. We're looking for the person who will help define how this practice gets built.
Most of your work will center on Notion Workers, Notion's hosted runtime for custom code: small Node/TypeScript programs that sync external data on a schedule, handle webhooks, and give Notion's agents new tools. You'll also write custom code for integrations and automations beyond Workers, connecting the systems our clients already use and handling the tasks native Notion can't. (New to Notion Workers? Here's Notion's overview.) You'll own these end to end, from spec to ship, then maintaining, debugging, and keeping them healthy for as long as they're live.
What You'll Do
What We're Looking For
This role is built around AI. Notion Workers are designed to be built with AI coding agents like Claude Code and Codex, and you'll lean on that tooling every day. That's exactly why the bar is on you: when AI writes the code, the engineer's judgment is what keeps it reliable, maintainable, and correct. We want the human who keeps the quality bar high, not a portfolio an agent generated unsupervised.