AI Engineer

About the role We are looking for someone who can bridge AI engineering and production operations. The candidate can take an AI prototype, implementing governance and testing, deploying it at scale, and continuously improving model performance while keeping the architecture flexible and independent of any single AI provider. This role is responsible for building enterprise-grade GenAI solutions that scale securely across the organisation, bridging AI engineering and production operations to take AI use cases from prototype to fully governed, production-ready applications. Stack includes Snowflake Cortex and AWS Bedrock, with a leading foundation model as the primary LLM provider.

KEY RESPONSIBILITIES

• Transition AI use cases from proof-of-concept notebooks into production-ready applications.
• Build evaluation frameworks to measure model quality, including automated regression testing and performance benchmarking.
• Implement AI safety guardrails to minimise hallucinations, ensure responsible AI usage, and maintain governance.
• Design and maintain Retrieval-Augmented Generation (RAG) pipelines.
• Develop a vendor-agnostic semantic layer, enabling applications to switch between LLM providers without significant redevelopment — avoiding vendor lock-in.

TECHNICAL SKILLS REQUIRED

• Strong proficiency in Python for AI application development and orchestration.
• Experience with Git and modern software development workflows (branching, pull requests, CI/CD).
• Hands-on experience with LLM optimisation, including prompt engineering, evaluation, tuning, and model selection.
• Understanding of vector databases, embeddings, RAG architectures, and enterprise AI deployment patterns.
• Note: while coding remains important, emphasis is increasingly on designing, evaluating, and optimising AI systems, as AI-assisted development tools reduce the need for extensive manual coding.

EXPERIENCE EXPECTATIONS

• 5+ years overall software / ML engineering experience, providing the systems, testing, and architecture discipline the role requires.
• 1–2+ years of hands-on, production (not just POC) LLM/GenAI experience — ideally including Bedrock, Snowflake Cortex, or a comparable enterprise AI platform.