Staff AI Engineer

Staff AI Engineer - Platform AI

London

4 days per week onsite

About the Company

Our client is a fast-growing B2B fintech building an intelligent banking platform designed for the next generation of financial services. The business is now expanding its AI capability and is looking for a senior engineer to help define how agentic AI is embedded into a complex financial technology platform from the ground up.

About the Role

We are looking for a Staff AI Engineer – Platform AI to provide technical leadership for the agentic layer of a major new platform. You will design how AI agents are authored, deployed, governed, executed and observed within a complex system combining data, rules, workflows and reporting. This is a hands-on architecture and engineering role. You will help define the control plane and runtime for production AI agents while ensuring autonomy, security, evaluation and governance are designed into the platform from the outset.

What You'll Do

• Design the agentic control plane for a new platform, including how

agents are authored, deployed, governed, executed and observed.

• Define agent registries, authority boundaries, autonomy levels and tool bindings.
• Design how agents integrate into versioned platform releases and the wider system architecture. •

Own

the evaluation and analytics layer for agent behaviour, including trace capture, hallucination detection, containment measurement, cost accounting and regression testing.

• Contribute to system-level design and architecture decisions.
• Ensure

AI systems are scalable, reliable, secure

and appropriate for a regulated financial environment.

• Work closely with core platform architecture and engineering teams to ensure agentic and deterministic capabilities operate as one cohesive system.
• Help establish the architectural foundations that will govern how agents operate across the platform as it scales.

What We're Looking For

• Extensive

AI/ML

or software engineering experience.

• Demonstrable experience designing and

shipping agentic systems into production

, rather than only building prototypes.

• Experience architecting agent governance for high-stakes environments, including guardrails, autonomy levels, kill switches and human-in-the-loop approval.
• Strong understanding of agent runtimes, including sandboxing, isolation, tool invocation and multi-agent orchestration.
• Experience with

Model Context Protocol (MCP)

or similar approaches to agent/tool integration.

• Experience building or operating evaluation and analytics frameworks for agent behaviour, including tracing, hallucination detection and regression testing.
• Experience designing systems combining deterministic, rules-based processing with agent-based reasoning.
• Strong Python and SQL skills.
• Hands-on experience with

LLM or agent frameworks

such as LangChain or similar.

• Experience with AWS and container technologies such as Docker and Kubernetes.
• Understanding of identity and authorisation for non-human or agent principals, including delegated authority.
• A data-driven approach to defining and measuring agent performance.
• Strong communication and collaboration skills.

Additional beneficial experience

• Agent-to-agent or tool-integration protocols at platform scale.
• AI/LLM evaluation frameworks and fine-tuning techniques.
• Graduated autonomy models and agent-as-principal identity patterns.
• Prompt-injection, jailbreak defence and other agent-specific security controls.
• Persona- or scenario-based simulation for testing agent behaviour.
• Hybrid architectures where deterministic processing and LLM-based reasoning operate within the same system.
• Token, spend or execution-budget governance for autonomous agents.
• Security and privacy within AI workflows, including encryption, anonymisation and compliance.
• Building an AI or agent capability from a blank sheet, spanning architecture, governance and production delivery.

You do not need to meet every requirement. We are particularly interested in engineers who have built sophisticated agentic systems and want to shape how AI operates within complex, regulated technology platforms.

Please note

UK visa sponsorship is not available for this position. Candidates must already have the right to work in the UK without sponsorship.