AI Engineer
Join a major organisation as it turns experimental AI into reliable production components. You will build agents, retrieval solutions, model integrations, and observability foundations that help teams ship trustworthy AI use cases.
CONTEXT
Our client, a major player in its sector, is entering a phase where artificial intelligence moves from isolated experiments to a trusted operational capability. The organisation has already mapped high-value use cases and now needs a hands-on AI Engineer to turn those ambitions into production-grade components: agents that can reason and act, retrieval layers that surface the right knowledge, model integrations that connect with existing systems, observability that makes AI behaviour understandable, and platform building blocks that other teams can reuse. The challenge is not only technical; it is about creating a safe, observable and maintainable AI foundation that can be adopted across multiple teams without fragmenting into one-off prototypes.
ACTIVITIES
- Design, build, test, and maintain reusable AI technical components, including agent workflows, retrieval pipelines, model integrations, and platform services.
- Implement observability and evaluation mechanisms so that AI behaviour, quality, latency, cost, and failure modes are measurable and actionable.
- Partner with product and engineering teams to translate concrete use cases into robust, secure, and maintainable AI solutions.
- Contribute to platform building blocks, documentation, and internal standards that accelerate adoption while reducing duplication and risk.
- Support teams during integration and experimentation, helping them move from proof of concept to production readiness.
- Investigate emerging AI patterns, tools, and model capabilities, then package the most promising ones into reliable components.
EXPECTED ACHIEVEMENTS
- A growing catalogue of production-ready AI components that teams can compose rather than rebuild.
- Faster and safer delivery of AI use cases, supported by clear observability, evaluation, and integration patterns.
- Stronger technical confidence across teams, with fewer fragile prototypes and more maintainable AI products.
- A measurable improvement in the reliability, traceability, and operational control of AI-powered services.
ENVIRONMENT
- Languages and frameworks such as Python and modern AI/LLM libraries.
- Agentic patterns, retrieval-augmented generation, vector search, model APIs, and integration middleware.
- Observability, evaluation, testing, and monitoring tooling adapted to AI workloads.
- Cloud-native engineering practices, API-first design, CI/CD, and collaborative agile delivery.