Principal AI & Agentic Tech, Lloyds

• Deep, hands-on experience with AI and Agentic AI at scale, with a practical understanding of what works in real-world environments, where limits sit, and how issues emerge once systems are live
• Strong, current expertise in AI and Agentic AI, with practical breadth across models, agentic patterns, and end-to-end technology stacks
• Credibility with technical specialists and senior leaders, with the ability to cut through complexity and apply sound technical judgement when decisions carry material risk
• Motivated by technical responsibility rather than span of control, and comfortable acting as the trusted authority on complex, high-stakes questions where getting it right matters
• Strong judgement on agentic capability and limits, including how these limits emerge as systems move from pilot to scaled, live operation
• Proven ability to influence AI technology approaches and trade-offs, including model selection, reuse vs bespoke approaches, and degrees of autonomy
• Skilled at focusing technical discussions on what matters most, bringing clarity to decisions that impact scale, safety, and long-term sustainability
• Demonstrated ability to steer difficult decisions through insight and evidence, particularly where technical risk, long-term cost, or operational viability is at stake
• A pragmatic engineering mindset, able to simplify for scale, challenge over-engineering, and avoid unnecessary complexity or fragility

Desirable

• This role brings deep, lived technology expertise to the Group’s AI and Agentic AI agenda, knowledge gained not at a conceptual level, but through sustained proximity to complex systems in real environments
• You understand how agentic systems work across the full technology ecosystem from data, APIs and infrastructure to modelling, decisioning, and the consumption layer
• You are genuinely excited by technology, continuously horizon scanning, and able to go deep across the breadth of agentic capabilities, from build and evaluation through to control frameworks, model selection, and how agentic fits within the wider decisioning and data science ecosystem
• You convert business requirements into AI and agentic technology requirements, identify where current capabilities fall short, and drive the backlog of what needs to be built or brought in
• Senior stakeholders and domain teams rely on you as the go-to technical authority – advising on complex, high-stakes agentic initiatives and ensuring the organisations technology choices are grounded in how agentic systems actually behave at scale
• You’ll bring a strong understanding of how agentic systems operate across the full technology ecosystem — from data, APIs and infrastructure through to modelling, decisioning and the consumption layer
• You’ll be genuinely excited by technology, continuously horizon scanning, and comfortable going deep across the breadth of agentic capabilities — from build and evaluation through to control frameworks, model selection, and how agentic fits within the wider decisioning and data science landscape
• You’ll turn business requirements into AI and agentic technology solutions, spotting where current capabilities fall short and driving what needs to be built or brought in
• As a trusted technical authority, you’ll work closely with senior stakeholders and domain teams, advising on complex, high-stakes initiatives and ensuring technology decisions are grounded in how agentic systems truly behave at scale
• End-to-end technical clarity for Agentic AI — define what “good” looks like across data, integration/APIs, security, infrastructure, models, orchestration, and the consumption layer
• A prioritised Agentic AI capability backlog for the domain, based on a forward-looking assessment of capability gaps and business impact
• Clear perspective on the agentic technology landscape — track new tools, models, frameworks, and patterns, and recommend how they should be adopted and integrated with existing capabilities
• Higher-quality, better-grounded decisions on Agentic AI, supported by expert challenge on autonomy, model choice, orchestration, controls, and scalability before commitments are locked in
• Stronger technical and architectural foundations for scaling Agentic AI, reducing the risk of fragile, one-off solutions that fail when reused or operated at domain scale
• Greater coherence across Agentic AI initiatives within the domain, avoiding fragmented solutions, duplicated effort, and incompatible assumptions
• Confident and responsible Agentic AI adoption, with safety, explainability, and operational robustness embedded as core design criteria

• End Date: Thursday 04 June 2026