1 day ago
London, United KingdomSenior
Responsibilities
- Design, build, and operate foundation-model, retrieval, and workflow services for operational product use cases.
- Build retrieval and context pipelines, structured outputs, evidence references, and citation verification.
- Implement multi-step workflows with human review, approval, escalation, and safe-failure or abstention paths.
- Create evaluation datasets, automated tests, regression suites, and monitoring for task success, retrieval quality, source support, robustness, latency, cost, and regression.
- Apply secure engineering controls for prompt injection, unsafe tool use, access-control failure, data leakage, unsupported claims, retrieval failure, and model or configuration degradation.
- Collaborate with Platform Engineering on release, runtime, observability, support, and operational concerns.
- Review supplier code and artefacts, maintain documentation, and transfer knowledge into permanent FCA ownership.
- Translate stakeholder requirements into scalable, maintainable, and robust AI solutions.
Requirements
- Strong Python and production software engineering experience, including APIs, automated testing, and cloud-native services.
- Practical experience building and operating production applications using foundation models and LLM APIs, with strong prompt-engineering knowledge.
- Experience with retrieval and embedding pipelines, semantic search, or equivalent AI data pipelines.
- Experience evaluating and monitoring AI systems using golden datasets, regression suites, retrieval and citation quality measures, and human or automated evaluation.
- Experience designing evidence-grounded, human-authorized workflows with explicit failure, abstention, and escalation behaviour.
- Experience with data modelling, schema definition, and validation frameworks.
- Working knowledge of AI failure modes and appropriate controls, including hallucination, retrieval drift, context failure, injection, and degradation.
- Experience with version control, operational telemetry, secure engineering, access control, data-leakage prevention, and safe tool use.
- Ability to communicate technical trade-offs, review supplier work, transfer knowledge, and translate non-technical business requirements into scalable solutions.
- AWS and Bedrock or comparable cloud and foundation-model platform experience is particularly valuable.
- Experience with OCR or multimodal foundation models, agent or DAG orchestration, event-driven systems, graph-enhanced retrieval, temporal or provenance models, and S3 metadata or annotation approaches is particularly valuable.
- Experience in regulatory technology, financial services, or other controlled operational environments is particularly valuable.
Benefits
- Build production-grade AI systems embedded in FCA regulatory workflows rather than standalone proofs of concept.
- Gain hands-on experience with retrieval-augmented generation, agentic workflows, orchestration, evaluation frameworks, provenance, observability, and secure AI system design.
- Help shape the safe and responsible application of AI within a major financial regulator.
- Work alongside engineers, architects, product leads, and domain specialists on AI capabilities supporting Authorisations, Supervision, Enforcement, and AML activities.
Categories
About Cognizant
Cognizant is a public IT services and consulting firm that designs, builds, and runs enterprise technology, including digital engineering, cloud modernization, data/AI, and managed services. It sells consulting, systems integration, and outsourcing on multi-year engagements to large enterprises in healthcare, banking, retail, communications, and manufacturing. Founded in 1994 and headquartered in Teaneck, New Jersey, Cognizant is NASDAQ-listed (CTSH) and a Fortune 500 company.