Principal AI Engineer
gsk
Job Description
- Design, develop, test and implement AI-enabled solutions, agents and automation capabilities supporting the MES Development Team.
- Work with the MES Automation Manager to translate the MES AI and automation strategy into prototypes, (Minimum Viable Product) MVPs and scalable production solutions.
- Lead the technical implementation and adoption of AI within the development team, coordinating AI engineering activities and coaching developers in AI-enabled ways of working.
- Develop reusable AI agents, services, integrations, (Application Programming Interfaces) APIs, automation components and AI-ready knowledge solutions applicable across MES development.
- Apply prompt and context engineering, retrieval/RAG (Retrieval-Augmented Generation), agent orchestration and tool integration to build reliable and effective AI solutions.
- Define and execute evaluation, testing, acceptance criteria and monitoring for AI-generated outputs and agent behavior, and implement appropriate human oversight, traceability, logging, access controls and exception handling for AI-enabled solutions.
- Apply professional software engineering practices using technologies such as Python, C#/.NET, SQL, PowerShell, (Representational State Transfer) REST APIs, JSON (JavaScript Object Notation)/XML (eXtensible Markup Language), Git/GitHub, automated testing and code review.
- Collaborate with Architecture, Product, MES Engineering, Infrastructure and other GSK AI and technology teams to align solutions with enterprise platforms, standards and reusable AI capabilities.
- Integrate AI capabilities with approved MES, engineering, knowledge and enterprise data sources, ensuring secure and controlled access, and ensure AI solutions comply with applicable Responsible AI, cybersecurity, data privacy, data integrity, (Good Practice ) GxP, validation and engineering requirements.
- Drive reuse, standardization and continuous improvement of AI engineering practices, components and knowledge across the MES Development Team.
- Use AI coding assistants and agent workflows as the default way of building, testing and documenting, and validates every output before it is committed or released.
- Deliver and support AI solutions across the full lifecycle, including (Continuous Integration / Continuous Deployment) CI/CD, release management, monitoring, incident resolution, and operational support, ensuring >99.5% availability, (Service Level Agreement) SLA compliance, and solution reliability. Drive AI-enabled capabilities that improve MES developer productivity, operational efficiency, and data-driven decision-making, with success measured through adoption, user satisfaction, delivery efficiency, incident reduction, and realized business value.