AI System Engineer

The role

We are looking for a Senior AI Systems Engineer to build and operate the intelligence layer of our AI products. This is a software-engineering-led role, not a prompt-development role: you will make model, retrieval, tool and orchestration behaviour testable, evidence-grounded, observable and reliable in production.

You will own the implementation of the product’s AI systems, including model integration, retrieval and context assembly, multi-step workflows, evaluation instrumentation, provenance mechanics and failure handling.

What success looks like

· AI outputs are grounded in data, and their supporting sources can be inspected wherever the risk tier requires it.

· Task, retrieval, citation and workflow quality are measured through repeatable evaluation and regression suites, with hooks exposed for independent evaluation.

· Prompts, models, orchestration logic, datasets, configurations and evaluation assets are versioned production artefacts.

· Known failure modes have explicit controls, abstention or escalation behaviour, and observable operational signals.

What you’ll do

· Design, build and operate foundation-model, retrieval and workflow services that solve the product’s operational problem.

· Build context and retrieval pipelines, structured outputs, evidence references and citation verification appropriate to the use case.

· Implement multi-step workflows with explicit human review, approval, escalation and safe-failure (abstention) paths.

· Work with stakeholders to create evaluation datasets and automated tests for task success, retrieval quality, source support, robustness, latency, cost and regression.

· Uphold security-conscious coding standards, building robustness against prompt injection, unsafe tool use, access-control failure, data leakage, unsupported claims, retrieval failure and model or configuration degradation, using appropriate third-party tooling.

· Work with Platform Engineering on CI/CD, release, runtime, observability and support, and with the Technical Lead on architecture and engineering quality.

· Review supplier code and artefacts, maintain documentation and ensure knowledge transfer into permanent FCA ownership.

What you’ll bring

Proven technical experience

· Strong Python and production software engineering, including APIs, automated testing and cloud-native services.

· Practical experience building and operating production applications using foundation models and LLM APIs, with a strong understanding of prompt engineering techniques and best practices.

· Experience with retrieval and embedding pipelines, semantic search or equivalent AI data pipelines.

Role-specific know-how, skills and behaviours

· Evaluation and monitoring of AI systems as a practice (golden datasets, regression suites, retrieval and citation quality measures, human and automated evaluation techniques) rather than ad hoc demonstration.

· Designing evidence-grounded, human-authorised workflows with explicit failure, abstention and escalation behaviour.

· Experience with data modelling, schema definition and validation frameworks.

· Working knowledge of AI failure modes (hallucination, retrieval drift, context failure, injection, degradation) and their controls.

· CI/CD, version control and operational telemetry, with prompts, configurations, datasets and evaluation assets treated as versioned production artefacts.

· Secure engineering: access control, data-leakage prevention and safe tool use.

· Communicating technical trade-offs within a multidisciplinary pod, reviewing supplier work and transferring knowledge.

· Translating non-technical business requirements from stakeholders into scalable, maintainable, robust solutions.

Particularly valuable

· AWS and Bedrock, or comparable cloud and foundation-model platforms.

· Extracting and structuring information from unstructured documents using OCR or multimodal foundation models.

· Agent orchestration, DAG orchestration, event-driven systems, graph-enhanced retrieval, or temporal and provenance models, including S3 metadata and annotation approaches.

· Regulatory technology, financial services or other controlled operational environments.

Why you’ll love this role

· Build production-grade AI systems that solve real regulatory problems, embedded in FCA operational workflows rather than standalone proofs of concept or demonstrations.

· Get hands-on experience at the forefront of enterprise AI engineering, including retrieval-augmented generation (RAG), agentic workflows, orchestration, evaluation frameworks, provenance, observability and secure AI system design.

· Help shape how a major financial regulator applies AI safely and responsibly, building systems that are evidence-grounded, measurable, auditable and designed to support meaningful human judgement.

· Tackle complex engineering challenges that go well beyond model integration, including retrieval quality, failure handling, evaluation, governance, security, operational resilience and long-term maintainability.

· Work alongside experienced engineers, architects, product leads and domain specialists while contributing to a growing AI capability intended to have impact across Authorisations, Supervision, Enforcement and AML activities.