AI Foundry Engineer: Copilot & LLM Orchestration

AI Developer
6 Month contract initially
Based: 2-3 Days onsite in Leeds/Bradford
Rate - £Market rates p/d via Umbrella

We have a great opportunity with a world leading organisation where you will be provided with all of the support and development to succeed. A progressive organisation where you can really make a difference. We have a great opportunity for an AI Developer on a long term program of work.

Key Responsibilities:

* Design, develop and maintain Copilot solutions, intelligent agents, plugins, connectors and LLM workflows (e.g., Copilot Studio).

* Build scalable components (prompt orchestration, retrieval layers, automation flows, model interfaces, validation pipelines).

* Integrate with enterprise systems/APIs/data platforms, ensuring security, resilience and architecture alignment.

* Conduct rapid prototyping to validate feasibility, model behaviour, UX and performance.

* Implement secure by design and responsible AI practices (guardrails, controls, monitoring, auditability).

* Develop/optimise RAG components, embeddings, vector queries and metadata strategies for accuracy/reliability.

* Implement observability: logging, telemetry and LLM monitoring for quality and incident triage.

* Create reusable assets (prompt libraries, agent templates, connectors, test harnesses) and documentation.

* Translate design artefacts into build ready specifications and aligned solution designs.

* Co define test strategies and model performance thresholds with the AI Test Lead.

* Contribute to cross functional design/architecture reviews and standards evolution.

* Mentor colleagues and enable pro /low /no code teams to adopt AI safely.

* Ensure responsible AI principles (e.g., transparency, explainability, ISO
42001) are incorporated into all development.

* Provide insight to support business cases, investment decisions, risk assessments, and prioritisation discussions at AI governance forums.

* Collaborate with teams to ensure all AI development work is implementable, sustainable and aligned to enterprise architecture.

* Maintain a library of development artefacts, patterns and re usable assets to support repeatability and uplift maturity across the AI Foundry.

* Managing escalations supporting the wider Data & AI Leadership team.

Key

Skills & Experience:

Hands-on experience with Copilot. Ideally, candidates should have a strong background in Data Engineering or broader Engineering disciplines. It is essential that all candidates demonstrate solid, hands-on experience within the Microsoft ecosystem, specifically:

* Building and integrating agents within Copilot Studio.

* Leveraging Power Apps, Dataverse, Fabric, and Azure.

* Utilizing Foundry environments.

* Practical experience integrating agents, APIs, and associated components within these specific platforms.

* Higher education qualification (or equivalent experience) in Ethics, Law, Risk Management, Social Sciences, Data/Computer Science or relevant field

* Proven hands‑on experience building solutions using LLMs, AI APIs, Copilot Studio or agent frameworks.

* Strong understanding of vector databases, embeddings, RAG architectures and retrieval optimisation.

* Experience implementing secure‑by‑design practices including authentication, authorisation, data protection and auditability.

* Experience working within Microsoft Foundry‑style model and agent engineering, including LLM orchestration, RAG component optimisation, agent lifecycle management, versioning, monitoring, drift detection, and building reusable model/agent components governed under enterprise controls.

* Experience working with Microsoft Azure AI and cloud-native engineering, including integration with Azure AI services, secure deployment patterns, observability, telemetry, vector search and embeddings, and alignment with enterprise-grade cloud architectures used across the AI Foundry.

* Familiarity with Dev Ops, CI/CD, IaC, observability, monitoring and modern engineering pipelines.

* Ability to translate complex requirements or user needs into scalable, maintainable technical solutions.

* Ability to debug unexpected AI or model behaviour, including hallucinations, variability and reliability issues.

* Strong…