Senior AI Engineer →

DTU — Technical University of Denmark; København; Hybrid; DKK 900,000 - 1,500,000; Full time

Benefits offered by this job

Growth from within

Top equipment

Culture & coaching

Meaningful work

Ex-founders team

Team events & perks

Job summary

All Gravy is seeking an AI Engineer to build the AI layer of the platform, shipping agents that plan shifts, answer questions, and take real actions across 2,000+ sites in the Nordics, UK, and beyond. This build role requires shipping production agents and evaluating their performance while exploring new models and frameworks.

The role emphasizes strong TypeScript/Node.js production skills, curiosity, and a pragmatic approach to AI capabilities, with remote-friendly, European-time-zone alignment.

Qualifications

• Built agents shipped to real users and can discuss failures.
• Strong curiosity with side projects or experiments.
• Solid production engineering in TypeScript and Node.js.
• Experience designing scalable distributed systems.
• Fluent English, written and spoken.

Responsibilities

• Build agent harnesses. Design and ship agents that call internal and external tools to solve real customer problems, both interactively with users and autonomously in the background.
• Own the eval loop. Build the harnesses, datasets, and offline and online evals that tell us whether an agent got better or worse.
• Ship full stack. Take features end to end, from the manager and employee interfaces where schedules are built and shifts are picked up, through the backend services and data layers that power them.
• Run it in production. Own latency, cost, failure modes, and guardrails for the agents you ship.
• Investigate what is new. Try the new models, frameworks, and techniques as they land, build small things to find out where they actually hold up, and bring back a clear read on what we should adopt and what we should ignore.
• Work across teams. Partner with product and the other engineering teams so AI features fit the wider All Gravy ecosystem, and contribute to shared standards and architectural decisions.

Skills

React

TypeScript

Node.js

Full-stack

Prompt engineering

Distributed systems

English fluency

Tools

Vercel AI SDK

Mastra

Job description

All Gravy is the AI-native people operating platform for hospitality and retail.

We're modernising how frontline teams work, replacing scattered tools and outdated systems with a single app that’s actually built for the way people work & socialise today. Our AI Colleague gives staff instant answers & executes day-to-day tasks, freeing up managers to focus on people & customers instead of paperwork. We work with some of the most exciting brands in hospitality and retail, including Gail's, L'Osteria, Dishoom, Pizza Pilgrims, Ottolenghi, Q8 and 7-Eleven. Today over 2,000 restaurants and stores trust us to run their teams.

We've grown 10x in two years and are accelerating into new markets. Our team spans a wide range of nationalities and backgrounds, united by one mission: to create better work lives for the 80% of the world that do not sit behind a desk.

About the role

We are looking for an AI Engineer to build the AI layer of All Gravy: agents that plan shifts, answer manager questions, and take real actions across 2,000+ hospitality and retail locations in the Nordics and UK.

This is a build role, not a research role. You will ship agents that customers use in production, then own how well they work. We are looking for someone who has already built and shipped AI agents and who is relentlessly curious about what just became possible. The field moves faster than any roadmap, so part of the job is noticing the new capability before anyone asks you to, testing it, and telling us whether it changes what we should build.

What you will do

• Build agent harnesses. Design and ship agents that call internal and external tools to solve real customer problems, both interactively with users and autonomously in the background.
• Own the eval loop. Build the harnesses, datasets, and offline and online evals that tell us whether an agent got better or worse. You define what “working” means and you defend the number.
• Ship full stack. Take features end to end, from the manager and employee interfaces where schedules are built and shifts are picked up, through the backend services and data layers that power them.
• Run it in production. Own latency, cost, failure modes, and guardrails for the agents you ship. When one misbehaves at 6am on a Saturday, you are the person who understands why.
• Investigate what is new. Try the new models, frameworks, and techniques as they land, build small things to find out where they actually hold up, and bring back a clear read on what we should adopt and what we should ignore.
• Work across teams. Partner with product and the other engineering teams so AI features fit the wider All Gravy ecosystem, and contribute to shared standards and architectural decisions.

What we are looking for

• You have built agents that shipped to real users, not just demos. You can talk through what broke, what you changed, and what you would do differently.
• Genuine curiosity, evidenced by things you built. You have side projects, throwaway experiments, or strong opinions formed by actually running the thing rather than reading about it.
• Strong context and prompt engineering, including retrieval, tool design, and structured outputs. You know why an agent fails and it is usually not the model.
• Solid production engineering in TypeScript and Node.js, or the ability to get there fast from another typed language.
• Comfort reading and writing code, reviewing architecture, and working directly with infrastructure. You ship the fix rather than filing the ticket.
• Judgment about when an agent is the right answer and when a well placed SQL query or a boring deterministic rule is better.

You’ll thrive here if you

We're looking for people with great energy, a good sense of humour, and a drive for getting things over the line. You think elegantly but act pragmatically, love being part of a team, and never stop learning

• 5+ years of proven full-stack experience across React, React Native, TypeScript, and Node.js.
• Ability to design and architect highly available, fault-tolerant, and scalable distributed systems.
• Startup or scale-up background. You know what it means to move fast, make pragmatic calls, and revisit them when the time is right.
• Fluent English, spoken and written.
• Based within European time zones, whether from our Copenhagen or London office or remotely.

An advantage if you have experience in

• Vercel AI SDK
• Mastra or similar Agent framework

Why All Gravy

• Grow fast, for real. We promote from within, move quickly, and give you real responsibility from day one.
• Top equipment. You'll receive all new equipment to do your best work
• A culture that actually develops you. Honest feedback, active coaching, and the autonomy to experiment. We invest in you because your growth is our growth.
• Work that matters. Your work will directly impact the daily lives of millions of frontline workers.
• You’ll work with people who’ve built, scaled, and won. Ex-founders, operators, all here because they want to do it again, bigger.
• The good stuff too. Prime office location, team lunches, Friday bars, padel tournaments, wine tastings, offsites, top-tier equipment, and a fridge that’s always stocked. We work hard and celebrate harder.