Location: Onsite in Dublin office
Start Date: ASAP
Duration: 6-Month Contract
Daily Rate: €650 - €750 per day (Negotiable)
Summary
We are looking for a ML Engineer to join our clients’ AI research program. This is a unique, emerging role sitting at the intersection of insurance operations, product thinking, and agentic AI - ideal for someone who understands how insurance workflows actually run and is excited to reimagine them through autonomous AI agents. You'll be part of a small high impact team, working closely with AI Engineers and business stakeholders.
You'll own meaningful parts of the data platform end-to-end — from architecture to reliability — with the autonomy to make technical decisions and the expectation to deliver at scale without compromising data quality or governance.
Your work directly enables AI model training, business intelligence, and operational reporting: you're the foundation everyone builds on. AI tooling is a first-class part of how we structure the research project: you'll use it to accelerate data engineering work, and you'll be the quality bar on everything that is developed.
This role leans toward pricing and modelling depth and is deliberately broader: building agents, RAG pipelines and integrations end to end, and moving fluidly between model work, tooling and delivery as the project needs. It also contributes to making LLM behaviour reliable and measurable, but as a generalist.
Based in a Dublin hub, you’ll work within a team which has been established specifically to explore and validate a next‑generation AI‑native insurance operating model. The environment is research‑focused, collaborative, and experimental, with a clear separation from day‑to‑day business operations to allow deep technical investigation, structured analysis, and prototyping.
You’ll cooperate closely with experienced professionals, participating in real‑time enterprise innovation research in a regulated insurance setting while contributing meaningfully to a high‑impact feasibility study.
Responsibilities
- Design, build and ship agents end to end — from prompt/tooling design through RAG pipeline construction to integration with production systems.
- Build and maintain RAG pipelines and integrations across a range of data sources and services.
- Design evaluation metrics and evaluation loops for LLM agents, in collaboration with the Senior Data Scientist.
- Contribute to converting non-deterministic LLM solutions into deterministic, trustworthy ones through structured evaluation and guardrails.
- Move between model development, tooling/integration work and hands-on delivery as project priorities shift. This role is intentionally less specialised than the Senior Data Scientist.
- Build and integrate MCP-based tools and services to extend agent capability.
- Build monitoring/evaluation dashboards in code.
- Apply reinforcement learning techniques to model development where relevant.
- 7+ years in data science, or 3 years in AI-Software Engineering.
- 2+ years of direct LLM development experience (Anthropic, OpenAI, or
- Copilot/Azure OpenAI ecosystems).
- Experience designing evaluation metrics and loops for LLM agents.
- Experience turning non-deterministic LLM solutions into more deterministic, dependable ones.
- Python
- Agent development and prompt engineering, in any framework or SDK.
- MCP development and integration.
- Model metrics and evaluation.
- Reinforcement learning as applied to model development.
- Dashboards built in code.
- C# exposure for agents.
- Regulated sector experience.
