Senior AI Engineer, Agentic Platform

Role overview Own the development of customer-facing AI agents that help users make decisions and complete work. You’ll turn existing service APIs and semantic data into useful agent capabilities, then take each feature through design, deployment, evaluation, and iteration so that it is reliable and valuable in day-to-day use.

Responsibilities - Build production agents and tools for retrieval, multi-step workflows, and actions, using Amazon Bedrock, AgentCore, and MCP. - Translate customer needs and existing data services into appropriate endpoints, tools, and accessible metrics. - Shape how agent responses are delivered, including speed, confidence, explanation, and suggested next steps. - Create evaluation systems with labeled examples, groundedness and hallucination measures, and CI checks that catch regressions. - Instrument agent behavior, investigate failures, and improve quality, coverage, and cost as usage grows. - Track adoption, feedback, and task impact, and communicate results to product and business partners.

Requirements - 5+ years building production software or data-intensive systems, with strong Python skills and at least one LLM or agent feature shipped to users. - Hands-on AWS experience; familiarity with Amazon Bedrock and AgentCore is useful, though experience with comparable agent runtimes can transfer. - Practical experience designing agent orchestration, tools or functions, context assembly, and multi-step workflows. - Experience building retrieval over business data, with enough SQL and metrics fluency to assess whether answers use the right information. - Ownership of an evaluation process for a generative system, including labeled data, regression checks, and defensible measures of improvement. - Ability to assess agent behavior, measure user outcomes, and communicate clearly across technical and non-technical teams.

Nice to have Experience with LLM observability tools, semantic or metrics layers, operational SaaS domains such as field service or logistics, or technical leadership. Experience using coding agents in production work is also relevant.

Benefits and work setup Remote within the UK, with occasional team meetups. Applicants must already have UK work authorization; visa sponsorship is unavailable. Benefits include an employer-matched pension, flexible time off plus UK bank holidays, enhanced parental leave, learning resources, and a 24/7 employee assistance program.