Location: London, United Kingdom Department: Engineering
We are looking for a Backend Engineer who builds with AI at the centre of their craft — not as a novelty, but as a primary tool that shapes how they design, write, review, and ship code. If you've already integrated Claude Code, agentic coding tools, and MCP servers deeply into your day-to-day, and you have opinions about how to do it well, we want to talk.
You'll join a growing business with real engineering problems: a cloud-based property and content management platform with 11 integrations and more on the way, plus a new Booking & Request Management Platform with real-time data and predictive intelligence features. Your job is to ship meaningful backend work — fast — and to bring the rest of the engineering team along with you on AI-augmented practices.
We are language-agnostic. Our current stack is PHP (Symfony) and Node.js, but we'll happily take a strong engineer from Python, Go, Java, C#, Ruby, or elsewhere who can move quickly between languages — especially with AI tooling at their side. The bar is engineering judgement and AI fluency, not a specific language on your CV.
This role is open to EMEA based candidates.
You start a feature by thinking carefully about scope and interfaces, then orchestrate Claude Code (or equivalent) to scaffold and iterate.
You've written CLAUDE.md / AGENTS.md files that materially improve agent behaviour, and you know why they work.
You configure MCP servers to give agents the right context — internal docs, ticketing, databases, observability — and you have a view on what should and shouldn't be exposed.
You review AI-generated code with the same rigour as human-written code, and you can spot the failure modes that less experienced AI users miss.
You measure and talk about the impact: cycle time, defect rate, review burden, the things that actually matter.
You help other engineers level up — pairing, internal docs, lunch-and-learns, custom commands, whatever moves the needle.
AI-First Development: Lead by example in using Claude Code and agentic tooling for real production work. Establish patterns the team can follow.
Agent Configuration & MCP: Build and maintain agent instruction files, custom commands/slash-commands, sub-agents, and MCP server integrations that make AI tooling genuinely useful on our codebase.
AI Workflow Design: Define how AI-generated code enters our repos — review standards, test expectations, security checks, and the guardrails around it.
Backend Engineering: Ship real backend features and services. AI accelerates the work; it doesn't replace the engineering.
Code Quality & Review: Set the bar for what good looks like in an AI-augmented codebase. Peer-review human and AI-authored code with equal scrutiny.
Team Enablement: Upskill other engineers in AI tooling — share patterns, write internal guides, run sessions.
Collaboration: Partner with product, frontend, DevOps, and QA to align technology with business goals.
Backend Development: 4+ years of commercial backend experience in any modern backend language (PHP, Node.js/TypeScript, Python, Go, Java, C#, Ruby, etc.).
Deep AI Tooling Experience: Hands-on, demonstrable use of Claude Code, Cursor, or similar agentic coding tools as a primary part of your workflow. You should be able to walk us through specific things you've built or shipped using them.
Agent Configuration: Practical experience writing CLAUDE.md / AGENTS.md / equivalent files, custom commands, and sub-agent setups — and a clear view on what makes them effective.
MCP (Model Context Protocol): Working knowledge of MCP — connecting agents to internal tools, data sources, or services. Bonus points if you've built your own MCP server.
API Design: Designing and managing scalable, secure APIs.
Automated Testing: Strong testing instincts, especially important when reviewing AI-generated code.
Database Management: Solid grasp of MySQL/relational and NoSQL databases.
Communication: Strong English (B2+) and the ability to explain AI-augmented workflows to people who haven't used them yet.
Building MCP servers or other agent tooling from scratch.
Prompt engineering depth — eval harnesses, prompt versioning, structured outputs, tool-use design.
LLM application development — not just using AI to write code, but building features that incorporate LLMs.
Azure — our cloud platform.
CI/CD and DevOps — Azure DevOps, Docker, AKS, infrastructure-as-code.
Building services from scratch end-to-end.
Frontend awareness — React or similar.
We're being upfront: this role is for someone who already lives in AI-augmented development. We're not looking for someone who used Copilot for a sprint, or who read
