Artificial Intelligence Engineer in Manchester

Manchester (Hybrid - 3 days/week in office) | £60,000 - £65,000

The Opportunity

A household-name, global retail & hospitality business is standing up a brand-new AI function and looking for the person to build it.

There's no legacy estate to inherit and no technical debt to untangle.

Whatever architecture you put in place in the first few months becomes the standard the function runs on.

You’ll be one of four founding members of the team, acting as the sole hands‑on builder at the outset and working under the technical mentorship of the Automation Manager.

Growth is already planned for through 2027, as proven use cases scale and more data becomes available.

This is a role for someone who wants to work at the sharp end of applied AI inside a large, operationally complex business.

The Role

You’ll be central to how work gets done across the business’s global functions.

It’s not a conventional software engineering role, the emphasis is on designing, building and running practical AI agents, automations and orchestration across Copilot Studio, Azure AI Foundry and Claude, in service of productivity, decision-making, customer experience and team effectiveness.

As sole builder, you’ll own delivery end-to-end of architecture, platform choice, build, testing and handover, working across support functions, retail sites, markets, technology teams and third-party suppliers to turn business ideas into working solutions and take them from proof of concept into production.

You’ll get technical mentorship from the Automation Manager, but you’ll be expected to operate independently as the hands‑on architect and builder of the AI agent estate, with genuine ownership over decisions that will outlast the team’s current size.

• What You’ll Be Doing
• Discovery & proof of concept (the main focus early on)
• Partner directly with the business and Business Analysts to spot high-value automation and AI opportunities.
• Rapidly scope and build proofs of concept, testing feasibility and value before any production commitment.
• Decide and justify the right platform for each use case - Copilot Studio, Foundry or Claude.
• Help establish the documentation and knowledge repositories that governance and controls will be built on.
• Turning proven ideas into production
• Carry POCs with demonstrated business value through to production-grade agents or automations, with full testing and validation.
• Build Copilot Studio agents, conversation flows, fallback strategies, Adaptive Cards in Teams, wherever a low-code approach fits.
• Build pro-code Foundry agents in Python, using Azure Functions, custom skills and workflows where that’s the better fit.
• Develop Power Automate cloud flows with genuine orchestration: error handling, retries, long-running operations.
• Model and integrate data, connectors, REST APIs and MCP servers, with security and auditability designed in from the outset.
• Platform, architecture & governance (ongoing)
• Own the end-to-end architecture of the AI agent estate - making the platform and design calls, escalating trade-offs to the Automation Manager.
• Select the right knowledge and retrieval approach, managed RAG, owned RAG or live query for each use case.
• Build eval sets, retrieval-quality checks, guardrails and observability, and deploy through CI/CD.
• Operate within the environment strategy and DLP policies, ensuring retrieval is permission-aware by design.
• Document every agent and promotion decision to a standard that’s genuinely handover-ready.
• Build agents cost-effectively, keeping ROI and budget front of mind.
• What We’re Looking For

Hands‑on experience with both Microsoft Copilot Studio and Azure AI Foundry is a must, this is a dual‑platform role and you’ll need to be credible on each from day one.

• 2-6 years in applied AI, software or automation engineering, with a portfolio of real production work.
• A genuine, active interest in AI and automation, you’re already building agents, workflows or infra‑adjacent tools in your own time.
• Demonstrable architecture-level judgement: able to make and defend platform and design decisions unsupervised.
• Copilot Studio experience covering topic authoring, generative orchestration, connectors, knowledge sources and debugging.
• Python development with sound engineering habits; testing, version control, CI/CD.
• Hands‑on Azure experience, ideally including Function Apps, and comfort with cloud-native deployment.
• Practical experience running LLMs in production, including prompt orchestration, RAG and output evaluation.
• Confidence reasoning through Copilot-vs-Foundry and knowledge-design trade-offs and explaining the reasoning clearly.
• Strong written communication and structured problem‑solving; documentation needs to be good enough for someone else to pick up the work.
• Infrastructure, platform or cloud engineering background – comfort with production operations, monitoring and reliability, not just prototyping.
• Power Automate and the wider Power Platform, including Dataverse data modelling and security roles.
• Azure AI Foundry specifics: Agent Service, Prompt Flow, Azure AI Search, Semantic Kernel or the Microsoft Agent Framework.
• Machine learning or forecasting experience.
• Certifications AB-620 and AI-102/103.
• Why Apply
• Build a brand-new AI function from the ground up inside a large, operationally complex business.
• True greenfield: no legacy estate, no technical debt, and real ownership of the architecture from day one.
• A defined growth path, the team is set to expand through 2027 as use cases prove out.
• Hands‑on exposure across Copilot Studio, Azure AI Foundry and Claude.
• #J-18808-Ljbffr