Agentic Platform Engineer

Job summary

The Agentic Platform Engineer is tasked with building and operating an enterprise agentic engineering platform that enables AI agents to perform tasks across the technology estate. Responsibilities include designing autonomous workflows, integrating agents with enterprise platforms, and ensuring operational excellence through CI/CD and cloud infrastructure management using Python and cloud-native architectures.

Benefits

Qualifications

• Hands-on experience designing and operating production AI-agent or LLM-powered systems
• Strong software engineering capability in Python, including API design, automated testing, asynchronous processing, dependency management and production debugging
• Practical experience with agent orchestration, structured tool calling, workflow state, retrieval-augmented generation and human-in-the-loop patterns
• Experience building secure integrations with enterprise APIs, repositories, cloud services, CI/CD, observability or ITSM platforms
• Experience creating evaluation frameworks for agent quality, task completion, safety, reliability, latency and cost
• Strong understanding of agent security, including workload identity, least privilege, secrets management, sandboxing, approvals and audit trails
• Experience observing and operating LLM or agent workloads in production, including tracing, prompt/model/tool versioning, regression monitoring and incident response
• Strong AWS and cloud-native engineering experience, including Infrastructure as Code, containers or serverless services, networking, IAM and monitoring
• Experience with CI/CD and production ownership, with willingness and ability to participate in an on-call rota
• Ability to explain complex agent behaviour, risks and trade-offs clearly to engineering and non-engineering stakeholders

Responsibilities

• Design, build and operate the enterprise agentic platform, including orchestration, runtime services, model routing, memory, identity and policy enforcement
• Develop reusable agent frameworks, SDKs and patterns that enable engineering teams to build secure, scalable AI agents consistently
• Engineer autonomous and human-in-the-loop agent workflows that plan, execute, validate and escalate tasks within defined risk and governance controls
• Integrate agents securely with enterprise platforms, cloud services, APIs and knowledge sources using modern integration standards and Model Context Protocol (MCP)
• Build production-grade agent services using Python, cloud-native architectures, event-driven design, automation and Infrastructure as Code
• Implement robust evaluation, observability and continuous improvement capabilities, including testing, tracing, telemetry and performance optimisation
• Embed security, governance and responsible AI principles through least-privilege access, policy enforcement, auditability and data protection
• Own the reliability and operational excellence of the agentic platform, including CI/CD, cloud infrastructure, monitoring, incident response and production support
• Collaborate with Engineering, Security, Architecture, Data and Risk teams to define standards, document best practices and support adoption of the platform

Skills

Relocation