AI Engineer

Build and ship AI systems that hold up in production.

You will build agentic and LLM features, integrate them into our pension and advisory products, evaluate them, and get them running reliably in systems people use every day.

About the role

We are looking for an AI Engineer to join our newly established AI team, working in our Life and Advisor products. This is a strong hands‑on building role: where our architect sets the overall system design, patterns, and standards, you take real ownership of building those features, evaluating them, integrating them into our Java products, and getting them working reliably in production.

You will collaborate closely with senior engineers and our AI Lead, and you will be part of an agile team with a flat structure, where collaboration, curiosity, and continuous learning are central to the culture. There is plenty of room to grow your scope over time.

This role is delivery‑focused and expects someone who already knows how to build AI systems well. It suits someone who has shipped real AI features, is comfortable with agent orchestration and evaluation, and wants to deepen their craft alongside a strong team.

Why this role matters

AI only creates value here when it is engineered well and integrated into the products millions of people rely on. The features you build will sit inside pension and advisory systems used by Europe’s largest providers, including PensionDanmark and APG, supporting more than 8 million members and over €800bn in assets. Your work directly affects the reliability, security, and accuracy of systems people depend on for their long‑term financial security which is why building AI that holds up in production truly matters.

What you’ll do

You will build, evaluate, and ship AI features in and around our products, taking problems from design through to deployed, reliable features, working within the architecture and patterns the team has set.

In practice, you will:

• Build agentic workflows: implement single‑ and multi‑agent systems, orchestration, tool use, and knowledge‑base integrations for reliable production use.
• Integrate AI with our systems: connect models and agents to our Java services, internal APIs, and data sources.
• Build and maintain RAG data flows: ingest, chunk, index, and retrieve from internal structured and unstructured data so features have the right context to work with.
• Evaluate what you build: write and run evals, use the team’s evaluation framework to catch regressions, and push relevance, faithfulness, latency, and cost in the features you own.
• Ship to production and keep it healthy: refine prompts, test thoroughly, handle failure cases, deploy, and monitor cost and performance once features are live.
• Collaborate and grow: work day to day with the AI Lead, senior developers, and product teams, and take on more design responsibility as you build trust.

What we’re looking for

You bring a strong engineering foundation and real proficiency building AI systems and getting them into production. You don’t need deep expertise in every area below, but we do expect genuine hands‑on strength in building, orchestrating, and evaluating AI.

Required

• A master’s degree in computer science or a similar field (preferably with an AI focus), or equivalent practical experience.
• Around 3–5 years of hands‑on experience working with AI and Large Language Models.
• Proficiency in a statically typed, object‑oriented language, Java or C# (we build in Java), or the ability to get up to speed quickly. Python is welcome for demos and experiments, but it is not our production language.
• Strong, demonstrated proficiency building AI systems, including hands‑on experience with one or more agent orchestration frameworks (e.g. LangChain/LangGraph, Microsoft Agents, Claude Agent SDK, AutoGen, Semantic Kernel).
• Practical experience with LLM and agent evaluation: building or using eval sets, and the metrics that distinguish a good system from a confident‑but‑wrong one.
• A track record of shipping AI systems to production, not just prototypes or notebooks, and keeping them working once live.
• Practical experience with Retrieval Augmented Generation (RAG) and connecting models to data, plus prompt strategy and an awareness of LLM cost and latency.

Nice to have

• Comfort across the stack, from backend services to the data layer.
• Experience with evaluation automation, observability, or CI gates for AI systems.
• Broader machine learning experience.
• Hands‑on experience building and integrating MCP servers.

Who you are

You are a strong, hands‑on builder who likes seeing features land in front of real users, and who takes pride in shipping AI that actually holds up rather than just demos well. You are pragmatic, finish what you start, and care about quality and reliability. You stay open to feedback, learn quickly, and are eager to take on more as you grow. You are curious, persistent, and result‑oriented.