Key Skills

Job Description

Senior AI Engineer – Data & LLM Pipelines – up to 90K – Utrecht / Hybrid Join us and help built the AI platform for grant writing! We do not offer visa sponsorship or transfer. Applying for grants is complex and time-consuming. With Grantific we want to change that and help researchers and companies do what they do best: focus on impactful research and innovation. That’s why we built an AI-powered platform that simplifies proposal writing. We are now extending it with a grant matching tool, to help our users find the funding they need. Underneath both sit hard language problems: long-form writing that stays faithful to its sources, retrieval across dense, multilingual funding documents and understanding what a call actually asks for. We work closely with Catalyze, a funding consultancy, so real grant experts are a desk away. You'll work across all of our AI features: the proposal writer, grant search, and the document processing that is essential for both features. You take a problem from idea to production and keep it running there. We ship small changes daily, release every week, and hold every AI feature to a clear quality bar: real data, frozen baselines, standard metrics and full observability. What you'll do Design the solution: turn a product problem into an approach: which model and data, what to measure and the simplest version that brings value. Build it end to end: from prompts, models and pipelines to the services the product calls, shipped in small, well-tested changes that reach users within days. Run it in production: instrument what you ship, watch the dashboards and alerts, read the logs and fix what breaks before anyone asks. Work across features: the agent that drafts and edits proposals with citations and guardrails; search over funding calls and the vocabularies that define them; parsing of long, messy documents in many languages. Build evaluation: with our grant experts (gold sets, faithfulness checks, ranking metrics) and make it part of how we release. Keep an eye on new models and techniques: judge which ones are worth the effort, say where we could do better and run the experiment that settles it. What we're looking for 4+ years shipping ML/NLP systems that real users rely on, end to end: design, build, deploy, monitor. Production LLM systems (retrieval, structured extraction, agents or long-form generation), and a concrete account of how you measured whether they worked. NLP foundations beyond LLMs: embeddings, retrieval and ranking, classification, multilingual text. Some experience modelling knowledge for search: taxonomies or controlled vocabularies, entity resolution, aligning data to existing classification schemes. Senior-level Python with sound habits: tests, small PRs, typed and maintainable code; at home with APIs, databases and cloud deployment. Comfortable releasing frequently and owning what you ship, with the monitoring and quick fixes that come with it. Proactive, independent and happy to switch between features; open to discuss features with non-technical colleagues and domain experts and help them if needed. Nice to have Pydantic AI or Logfire, knowledge graphs, long-document understanding, regulatory or public-sector text or a research background in NLP or information retrieval. Our stack Python and Pydantic AI, Gemini and Claude on Vertex AI, self-hosted open models on RunPod, Temporal for pipelines, Google Cloud, Logfire. None of this is set in stone. If you’ve worked with something better, bring it along. What we offer A small, ambitious team with real freedom in how you work, and plenty of room to grow as the company does. Real impact: you help some of Europe's most innovative companies and researchers get funded. Continuous growth & learning: We actively foster curiosity, research and skill development through courses, valuing the expertise you contribute back to our team. Hybrid, with three days a week together in our Utrecht office and occasionally in Catalyze office in Amsterdam. A competitive salary including holiday allowance, secondary benefits, a laptop and a travel card. For international hires, we support applications for the Dutch expat tax ruling where eligible. Interested!? Apply via LinkedIn.

Core Responsibilities

Design, build, deploy, and operate AI features for proposal writing, grant search, and multilingual document processing, including LLM pipelines, retrieval, and production services. Develop evaluation methods with grant experts, monitor production quality, resolve issues, and assess new models and techniques.

Requirements

Candidates need at least four years of experience shipping end-to-end ML or NLP systems and hands-on experience with production LLM systems, retrieval, extraction, agents, or long-form generation. They should have strong Python engineering practices, NLP foundations, some experience with taxonomies or controlled vocabularies, and the ability to own frequent releases and collaborate with non-technical experts.

Benefits

• Holiday Allowance
• Secondary Benefits
• Laptop
• Travel Card
• Dutch Expat Tax Ruling Support (Where Eligible)
• Professional Development