We are seeking a Lead AI Software Engineer in setting the engineering standard for how AI coding assistants such as Claude Code and frameworks are used safely and effectively across delivery teams.

This is a hands-on role. You will spend the majority of your time writing code, reviewing code, and shaping technical implementation. You will be expected to set the bar for quality, security, and operational readiness, and to lift the engineers around you through direct technical mentorship rather than through process or strategy.

What You'll Do

Hands-On Software Engineering

• Design, build, and ship production software across the AI delivery stack backend services, APIs, and integration layers to a banking-grade standard of quality, observability, and resilience.
• Set the technical pattern through reference implementations that other engineers can extend, including service scaffolding, integration adapters, and agent orchestration components.
• Lead code review across the team. Hold the line on testing discipline, secure coding practice, error handling, performance, and maintainability and coach engineers through the reasoning behind each call.
• Own non-functional engineering: logging, tracing, metrics, secrets management, dependency hygiene, and CI/CD pipeline quality. Make the path of least resistance the correct path.
• Drive technical refactoring and modernisation of existing services where AI-assisted delivery exposes structural debt that limits velocity or safety.

AI-Assisted Software Engineering

• Use Claude Code and equivalent AI coding assistants as a daily engineering tool. Build the prompts, sub-agents, slash commands, hooks, and workflow conventions that make them effective on real banking codebases.
• Pair AI-assisted delivery with the engineering discipline a regulated environment demands explicit human-in-the-loop checkpoints, deterministic test gates, traceable change history, and clear separation between AI-generated and human-authored code where audit requires it.
• Build internal tooling that wraps AI coding assistants for banking use: codebase-scoped configurations, redaction layers for sensitive data, repo-aware prompts, and review automation that enforces the team's quality bar.

Embedding AI-Assisted Delivery in Existing Engineering Squads

• Work directly inside existing engineering squads as a hands-on technical contributor pairing on real stories, writing code alongside their engineers, and integrating AI tools into their actual development workflow rather than running it as a separate programme.
• Build the reusable engineering assets that make adoption stick: repo templates, agent configurations, prompt libraries
• Run technical workshops and pairing sessions with squad engineers.
• Identify the engineering tasks where AI assistance has the highest return code review, test generation, documentation, refactoring, integration scaffolding, log analysis and build the tooling and prompts that make those wins repeatable.

Engineering Guardrails for AI Tooling in a Banking Environment

• Define and implement the engineering controls that govern AI coding assistant usage: which codebases and data are in scope, what AI-generated output requires human review before merge, and how AI-assisted commits are recorded for traceability.
• Build the technical enforcement of those controls repository configuration, branch protection, CI checks, prompt logging, and telemetry rather than relying on policy alone.
• Partner with Security, Risk, and Compliance engineering counterparts to ensure AI tooling integrates cleanly with existing controls around secrets, data classification, change management, and SDLC evidence.

Technical Mentorship

• Raise the engineering capability of the team through direct technical mentorship pairing, code review, design review, and worked examples not through abstract guidance.
• Author the internal technical playbooks, reference implementations, and engineering standards for AI-assisted development inside the organisation, and keep them current as the tooling evolves.
• Be the engineer other engineers come to when something is genuinely hard. Hold the bar on technical quality without becoming a bottleneck.

What We're Looking For:

Essential Engineering Experience

• 8+ years of hands-on software engineering experience, with a meaningful portion at principal or staff level shipping production systems that other engineers depend on.
• Deep proficiency in at least one modern backend language (e.g. Python, TypeScript/Node.js, Java, Go) and the ability to operate effectively in a polyglot codebase.
• Strong applied experience with API design, distributed systems, message-driven architectures, and integration with enterprise systems of record.
• Demonstrable experience designing for security, observability, and operational readiness in production not as an afterthought.
• Solid command of testing discipline across unit, integration, contract, and end-to-end levels, and of CI/CD pipelines that enforce it.
• Experience building software inside a regulated environment, or comparable evidence of working under audit, change control, and data protection constraints.

Ways of Working

• A senior individual contributor at heart most comfortable when the work is in code, in pull requests, and in technical conversations with other engineers.
• Able to translate complex AI engineering decisions into clear, plain technical writing for engineering leads, security partners, and audit-facing stakeholders.
• Pragmatic under ambiguity. Able to make defensible technical calls without waiting for perfect information, and willing to revisit them when evidence changes.
• Holds the engineering bar without becoming a blocker. Decisive in code review, generous in technical mentorship.