Data Scientist

AI / LLM Engineer (Copilot, Azure AI) Leeds (Hybrid – 3 days onsite) | Contract | Outside IR35 | Competitive Day Rate We’re looking for a hands-on AI / LLM Engineer to design and deliver enterprise-grade Generative AI and Copilot solutions within a modern cloud environment. This role sits at the intersection of AI engineering, architecture, and platform delivery, with a strong focus on building scalable, secure, and production-ready AI systems. What You’ll Be Doing • Design, build and deploy Copilot solutions, AI agents and LLM-powered applications • Develop end-to-end LLM workflows (prompt orchestration, automation, integrations) • Implement and optimise RAG pipelines, embeddings and vector search strategies • Integrate AI solutions with enterprise systems, APIs and data platforms • Build scalable and reusable components (prompt libraries, connectors, templates) • Apply secure-by-design and responsible AI practices (guardrails, monitoring, auditability) • Implement observability and monitoring (logging, telemetry, performance tracking) • Troubleshoot and improve model performance, reliability and hallucination issues • Collaborate with cross-functional teams across engineering, architecture and governance Key Skills & Experience • Proven experience building LLM / Generative AI solutions in production • Strong understanding of: • RAG (Retrieval-Augmented Generation) • Embeddings and vector databases • Prompt engineering and orchestration • Hands-on experience with: • Azure AI / Azure OpenAI • Copilot Studio or similar agent frameworks • Experience integrating with APIs and enterprise systems • Knowledge of secure development practices (authentication, data protection, compliance) • Familiarity with DevOps, CI/CD, monitoring and cloud-native architectures • Ability to debug AI behaviour (hallucinations, inconsistency, performance issues) Nice to Have • Experience in financial services or regulated environments • Exposure to AI governance, risk and compliance frameworks • Experience with agent frameworks (e.g. orchestration tools, Semantic Kernel, LangChain)