A public-sector IT department of around 150 staff is using generative AI and retrieval-based approaches to make document flows and knowledge access more useful for internal teams. This role exists to build, integrate and operate production-ready AI services that connect Python-based AI pipelines with the organisation's predominantly C#/.NET applications, with a strong focus on document processing, information extraction and reliable API integrations.
The mission
You will join a multidisciplinary delivery environment working in Agile SAFe, contributing technical ownership for AI features that support business processes such as intelligent assistance, automated classification and knowledge retrieval. The technical landscape includes on-premise deployments, containerised services, CI/CD pipelines and integration points with existing APIs and backend systems; solutions should emphasise explainability, bias control and operational reliability.
On a day-to-day basis you will translate use cases into production implementations: design retrieval and grounding pipelines, implement document intelligence for extraction and structured outputs, and expose functionality through stable API contracts that .NET teams can consume. You will also set up evaluation frameworks and monitoring for AI quality metrics, and work with architects, security and product owners to keep solutions maintainable and auditable.
Your responsibilities
- Deliver production-ready AI services that enable document processing, information extraction and intelligent assistance, prioritising reliability and measurable quality outcomes
- Design and implement retrieval-based solutions and grounding for LLM-driven flows to reduce hallucinations and improve factuality
- Integrate AI components with existing applications via well-defined API contracts and SDKs, collaborating closely with C#/.NET development teams
- Establish evaluation frameworks and test scenarios to measure accuracy, consistency, bias and grounding, then iterate on prompts, retrieval logic and output structures
- Implement CI/CD, versioning and containerisation practices to move experiments into stable, auditable production deployments
- Operate monitoring, logging and observability for AI services using tools and standards such as OpenTelemetry and application tracing to track latency, error rates and inference cost
Your profile
Essential skills
- Medior-level AI engineer, able to deliver production AI services and integrations, typically with around 3+ years of relevant experience
- Proficient in Python for model integration, retrieval pipelines and backend services
- Practical experience with C#/.NET to design API contracts and collaborate with existing application teams
- Hands-on with generative AI and LLM-based applications, including prompt engineering, retrieval/grounding and output structuring
- Strong background in document processing: information extraction, classification or knowledge unlocking and producing structured outputs
- Experience with CI/CD, containerisation and deployment of AI components, plus version control for models and services
- Competent with observability, logging and tracing for AI applications, familiar with OpenTelemetry or commercial APMs
- Able to define and run evaluation frameworks that track bias, hallucinations, relevance and reliability
- Clear communicator who can explain technical choices to mixed technical and non-technical stakeholders
Languages
- Dutch or French, CEFR C1
- Second national language, CEFR B1 (understanding and basic collaboration)