Join us as an AI Principal Engineer

• This is an opportunity for a driven senior engineer to take on an exciting new career challenge that combines the core skills of a software engineer with specialised knowledge in AI to build intelligent systems that can solve complex problems and drive innovation within NatWest
• You'll play a pivotal role in the Wealth AI centre of excellence (CoE) team in architecting, building and maintaining a highly extensible, scalable and maintainable platform that offers effective, reliable, and ethical AI solutions.
• You'll gain valuable stakeholder exposure as you build and leverage relationships, as well as the opportunity to hone your technical talents

• Designing, building, and maintaining scalable and robust AI infrastructure and platform services
• Responsible for the end-to-end delivery of services and features ensuring high quality and timely delivery
• Building tools and pipelines for prompt and context engineering, fine-tuning, evaluating, and monitoring of Large Language Models (LLMs)
• Collaborating with Machine Learning researchers, data scientists, and business journey product teams to understand user cases and deliver reusable platform capabilities
• Contributing to the development of platform UI components, libraries, and documentation to accelerate LLM adoption
• Working across the life cycle, from requirements analysis and design, through coding to testing, deployment and operations

• Strong programming skills in Python or Java
• Experience with cloud platforms such as, AWS, Azure, GCP and container orchestration, such as Kubernetes
• Experience working with vector databases and retrieval-augmented generation (RAG) architectures
• Experience building RESTful APIs for internal and external consumption
• Proven knowledge of web technologies and web security
• Experience with building and maintaining data pipelines with messaging systems such as Kafka and Active MQ
• DevOps knowledge including Docker, GitLab, ArgoCD, or similar tools for automation and deployment pipelines