The Role

We're hiring backend engineers to own our systems end-to-end - the data models, workflows, and APIs behind dispatch, ticketing, settlements, and integrations. You'll work with frontend, mobile, product, and our Forward Deployed Engineers, and you'll have meaningful input on how the system should behave for scalability and performance.

Our code has been ~100% AI-generated since March 2026. It's changed how the team works. We're building the software factory that turns customer, PM, and engineering input into safe deployable code. We expect the engineering team to be spinning up agents and managing the output rather than writing code by hand. While AI generates code here, engineers own the architecture, invariants, tests, review, security, and production behavior of everything we ship.

Stack

Rails API (Ruby), Postgres, REST. React on the web, Flutter on mobile. AWS.

What You'll Do

• Manage state machines across projects, orders, jobs, and loads.
• Operate a telematics layer with billions of rows tracking thousands of trucks across North America - live position, route history, mileage attribution.
• Design geofencing and geo-driven event processing that turns raw GPS into structured dispatch events.
• Build algorithmic dispatch and optimization, including automated load assignment.
• Run the payments stack that settles ~$1B/month in hauler payouts.
• Ship customer-facing agentic workflows on our MCP and LLM features.
• Build integrations with telematics providers, scale houses, ERPs, and other external systems.
• Shape our internal AI tooling - shared agent skills, automated review, prompt and skill infrastructure.

Requirements

• Strong Rails experience, or strong experience in a comparable backend framework and willingness to ramp fast.
• You think in systems - interfaces, schemas, invariants, failure modes - and you've shipped designs that held up as the product scaled past their original assumptions.
• You can model workflows where the source of truth is unreliable and the system still has to settle correctly.
• High ownership.
• Daily, opinionated use of AI coding tools.

Preferred

• You can debug production issues and explain root causes.
• Background in logistics, ops, payments, or fintech.

How We Work

We deploy continuously, review PRs the same day, and invest in correctness up front so on-call stays light.

AI tooling is part of the infrastructure. We maintain a shared internal skills library across Claude Code and Cursor, with namespaced agent skills the whole team uses for reviewing diffs and resolving PR feedback. We engineer the agent loop itself - skills, prompts, review pipelines. We invest in the dev environment and CI work that makes coding agents useful: devcontainers, fast feedback, pre-approved permissions, cloud agent support. Improving the SDLC is first-class engineering work.

Before You Apply

This isn't the right role if you want to:

• Write only CRUD endpoints.
• Work from fully-specified requirements.
• Treat AI coding tools as a side experiment.

How to Apply

Send something you've worked on where the backend had real complexity. A repo, a write-up, an architecture doc, a postmortem. One good thing beats a long resume.