Agentic AI Engineering Lead — Enterprise Agent Orchestration - SVP

The Department

Developer Engineering is a function of the CTO organization. Our mission is to make it easy and enjoyable for software engineering teams to go from a business idea to delivering an innovative product solution. We are committed to modernizing our toolchain, streamlining delivery processes, automating at scale, and embedding intelligent controls that help engineering teams ship with confidence and speed.

Within the Developer Engineering department, the Developer Services group is a dedicated expert team at the forefront of the everything-as-code agenda. We exist to deliver measurable reductions in process friction, manual effort, and human error — ensuring our policies, standards, and controls are codified, automated, and consistently applied across the organization.

We hold a unique mandate as part of a greenfield program to shape critical technical consensus at global scale, transforming how the firm's engineers build and apply controls across its technology landscape.

Developer Services is a cross-functional team of engineers, AI practitioners, data scientists, business analysts, and product managers — working together to engineer next-generation codified controls, build the platforms that power them, and drive enterprise adoption.

This is a rare greenfield / build-from-scratch opportunity to create an enterprise agentic AI capability — the agent orchestration layer, the guardrails, and the adoption model — inside one of the world's most highly regulated financial environments.

As the Agentic AI Engineering Lead, you will own the enterprise rollout of autonomous and semi-autonomous AI agents across the software delivery lifecycle: designing the multi-agent orchestration platform that plans, routes, executes, and audits agent work, and defining the control model that makes AI autonomy safe under financial services regulation. You will start from a blank page, set the technical strategy, lead a team of engineers, and partner with product, security, risk, controls, and platform teams to take agents from pilot to firm-wide production.

You will work hands-on with frontier Generative AI technologies — large language models (LLMs) such as GPT, Gemini, and Claude — through prompt engineering, context engineering, tool / function calling, retrieval-augmented generation (RAG), agentic planning loops, and evaluation harnesses that prove agent behaviour before production.

This is not just an engineering role. It is a leadership opportunity to define how agentic AI is built, governed, and scaled across a global bank.

• Architect and build the enterprise agent orchestration platform from the ground up: agent runtime, planner/executor loops, tool and MCP-style integration layer, memory and context management, state persistence, and human-in-the-loop checkpoints
• Design multi-agent patterns — supervisor/worker hierarchies, task decomposition, delegation, hand-off, and recovery — with deterministic fallbacks where autonomy is not appropriate
• Build the tool integration surface that lets agents act safely on real enterprise systems: source control, CI/CD, ticketing, service catalogues, data platforms, and internal APIs, each with scoped, least-privilege credentials
• Define the evaluation and observability layer: golden datasets, offline and online evals, regression gates on agent behaviour, prompt and model version control, token and cost governance, and end-to-end tracing of every agent decision and action
• Establish reusable agent patterns, SDKs, and templates so other engineering teams can build compliant agents without rebuilding the plumbing

• Own the enterprise rollout…