Job Title-AI Specialist (AI Foundry)

Client-Global Consulting Firm

Day Rate-£500-£550 (Outside IR35)

Duration-6 months (extendable)

Location-Leeds (Hybrid, 3 days per week)

• Translate complex concepts into clear guidance for executives and governance forums.
• Contribute to AI strategy, roadmap and horizon‑scanning, informing investment and risk posture.
• Provide insight to support business cases, investment decisions, risk assessments, and prioritisation discussions at AI governance forums.
• Lead enterprise‑wide assurance for high‑risk AI systems, conducting deep reviews, assessing model and agent behaviour under stress conditions, identifying systemic risks, and recommending design‑level mitigations to ensure solutions remain robust, ethical and compliant.
• Serve as the subject matter expert on AI technologies, including large language models (LLMs) and agentic AI, providing insights and recommendations to support business cases and investment decisions.
• Lead exploration and proof‑of‑concepts for multi‑agent, tool‑using and semi‑autonomous workflows.
• Define enterprise AI patterns for orchestration, prompt governance, vector search, memory and agent frameworks.
• Act as a key influencer and advisor to senior management and governance forums, providing expert guidance on AI-related decisions and strategies.
• Advise on model selection, prompting strategies, data approaches and trade‑offs for safe, high‑performing solutions.
• Shape and govern agentic AI patterns (memory, planning, tool use, autonomy constraints, escalation).
• Guide RAG implementations, vector databases, embedding pipelines and content governance models.
• Ensure strong data foundations (quality, lineage, metadata, privacy, access controls) for AI workloads.
• Drive cross‑functional technical AI enablement by developing advanced tooling, accelerators and evaluation frameworks, enabling teams to experiment safely, adopt best‑practice AI patterns, and scale AI delivery with higher consistency and lower technical debt.
• Define standards, guardrails, and reusable frameworks for Copilot/LLM/agentic solutions.