Turn this role into an interview — a resume and cover letter built around what this employer wants.
Amps is building the AI-native infrastructure layer for connected energy. We are looking for a backend-focused engineer to design and implement scalable services in TypeScript/Node.js, with a strong eye for distributed systems and reliability.
You’ll own the backend and contribute to the frontend in Next.js/React, collaborating across teams. Hybrid work in London, with freedom to shape architecture and tooling.
Amps is building the AI-native infrastructure layer for connected energy.
The grid is being rebuilt around millions of small, distributed assets - home batteries, solar inverters, heat pumps, EV chargers and electric vehicles. Each manufacturer ships its own API, its own auth model, its own quirks and its own idea of what "state" means. Anyone trying to build a real energy product ends up spending months on integrations before they write a single line of the thing they actually set out to build.
We remove that problem. Amps gives developers one REST API, typed SDKs, webhooks and a sandbox to read real-time telemetry from and issue control commands to devices across Tesla, Enphase, SolarEdge, Daikin, Mitsubishi, Wallbox, BMW and every other manufacturer our customers need.
Our customers are OEMs exposing device data and control at scale, energy software builders shipping customer‑facing products, and utilities operating distributed energy portfolios across grids and markets. If it has an API, we connect to it and we make it behave like one coherent system.
You'll work across the stack, with the centre of gravity on the backend: the services that talk to hundreds of third‑party device APIs, normalise wildly inconsistent data into a single model, and move commands out to physical hardware reliably and quickly.
This is a systems job as much as a product job. Devices go offline. Vendor APIs rate‑limit, time out, and change. Commands have real‑world consequences - a battery that doesn't charge when it should costs someone money. Getting this right means thinking carefully about queues, retries, idempotency, caching and back‑pressure, and then building the developer‑facing surface that hides all of it.
You'll also build the front‑end our customers use: dashboards, onboarding flows and developer tooling in Next.js and React.
You don\'t need an energy background to do well here. Several of the hardest problems are ordinary distributed systems problems wearing an unfamiliar hat. Winning is what counts.