What you'll be doing

• Own the AI layer of production systems for public sector clients, including agents, model integration, and backend services
• Build production agents using established architectural patterns with reliable tool calling, guardrails, and observability
• Select, adapt, and design evaluation methods for client-specific risks
• Apply context engineering, retrieval-augmented generation, model adaptation, or inference optimization
• Investigate system regressions, lead root-cause analysis, and implement preventative changes
• Lead project planning and manage client expectations on scope, risk, and delivery
• Create reusable tooling, evaluation components, or guidance for recurring client work
• Mentor engineers through code review, pairing, and delegated ownership

Requirements

• Backend engineering experience with a proven track record of deploying AI/ML systems into production and maintaining their reliability
• Responsibility for AI behaviour of a production system end to end, not just scoped components
• Hands-on evaluation design linking model behaviour to real-world outcomes to detect regressions
• Comfortable engaging directly with clients and explaining model behaviour and trade-offs to non-technical stakeholders
• Practical knowledge in handling regulated or sensitive data and designing for auditability; experience with sovereign or on-premise deployments is valuable
• Experience leading root-cause analysis on production AI or ML systems