AI/ML Engineer

• Demonstrable experience developing, deploying and operating machine learning and AI solutions in production environments, not proof of concept
• Hands on experience building LLM applications, including RAG, in Python with frameworks such as LangChain. Demonstrate a working understanding of why RAG solutions underperform and how to fix them
• Practical experience of vector databases and embeddings, such as Qdrant or Azure AI Search, and evaluating retrieval quality rather than assessing it by inspection
• Experience of monitoring live AI services - answer quality, drift, latency and cost - including attributing inference cost to use cases and tracing increases back to prompt or model change, using ai monitoring and inference gateway tooling
• Proven knowledge of Kubernetes or OpenShift and containerised deployment, with CI/CD in Gitlab or equivalent, secrets management, and monitoring with Prometheus, Grafana and Loki
• Experience implementing governance and guardrails for AI services in a regulated or security conscious environment 0 model admission and provenance, responsible AI controls and auditable of what was deployed and why

Salary: Competitive, depending on experience

We are seeking a Senior AI & ML Engineer to design, build and support production-grade AI solutions that deliver measurable business value. You will develop and deploy large language model (LLM) applications end-to-end, including retrieval logic, prompt engineering, orchestration workflows and supporting APIs.

Working closely with data scientists, engineers and business stakeholders, you will implement Retrieval-Augmented Generation (RAG) solutions, optimise data retrieval strategies and ensure secure access controls are enforced. You will define and maintain frameworks for evaluating answer quality, monitor model performance and drive continuous improvement through testing and governance.

The role includes taking experimental AI solutions into production, ensuring code quality, reproducibility and operational resilience. You will manage model and prompt lifecycles through MLflow, oversee model promotion and rollback processes, and deploy AI services to Azure and OpenShift using GitLab CI/CD pipelines.

You will also monitor live services for quality, drift, latency and cost, helping shape and evolve the organisation's AI ecosystem.

• Build and run production LLM applications end to end, from use case design through to live service - retrieval logic, prompt design, orchestration and the APIs behind them
• Design and implement RAG solutions, including chunking and embedding strategy, vector indexing and retrieval filtering that enforces user entitlements at the data layer
• Establish how answer quality is measured - define evaluation sets, baseline performance before release and re-evaluate after any change to prompt, model or document set
• Take experimental work from data scientists into production, including code review, refactoring for modularity, dependency handling and making it reproducible
• Own the model and prompt lifecycle in MLflow - tracking, registry, versioning, promotion and rollback. Ensure training and inference remain consistent
• Execute promotion of models and use cases into production serving, against a governance approved admission policy
• Deploy AI inference services into OpenShift and Azure with CI/CD pipeline in GitLab, comprehensive monitoring, logging and alerting, and secrets management
• Monitor live AI services into OpenShift and Azure for answer quality, drift, latency and cost, investigate regressions and trace cost increases to their cause
• Work with data scientists, data engineers, platform engineers and business stakeholders to industrialise AI use cases, and support the evolution of the AI ecosystem

Location: UK Wide. We offer a range of hybrid and flexible working arrangements. Please speak to your recruiter about the options for this particular role.