Unlocking decades of agricultural expertise with RAG technology
A custom-built intelligent platform enabling advisors to query over 100,000 historical documents in seconds.
DLV Advies
DLV Advies, a leading Dutch agricultural consultancy, required a way to make their vast archive of over 100,000 historical documents accessible to their advisors. The goal was to build a secure, high-performance knowledge base that allows users to ask complex questions in Dutch and receive cited, accurate answers. SevenLab developed a custom RAG (Retrieval-Augmented Generation) application to bridge the gap between decades of expertise and modern AI efficiency.
Advisors spent significant time manually searching through fragmented document archives consisting of PDFs, spreadsheets, and emails. With over 745,000 individual text chunks, standard keyword search was insufficient for finding specific precedents or comparing regional projects. The organisation needed a system that could understand agricultural context, handle massive data volumes, and provide verifiable answers with direct links to source files.
A look inside
What SevenLab built
SevenLab engineered a robust AI platform using a hybrid retrieval architecture that combines pgvector similarity search with trigram-based keyword matching. The system utilises Gemini Flash for high-speed reasoning and structured answer generation, supported by a custom Bun-based backend for real-time document processing and ingestion.
Hybrid search engine
Combines vector similarity and keyword matching to ensure pinpoint accuracy across 106,000 documents.
Verified citations
Every AI-generated response includes clickable sources that open the original document chunk for immediate verification.
Automated ingestion pipeline
A high-performance pipeline that handles bulk imports from external volumes with AI-driven deduplication and screening.
Agentic reasoning steps
Real-time transparency showing the AI's classification, search strategy, and data synthesis process.
Measurable business impact
The DLV Kennisbank has transformed how advisors access institutional knowledge, reducing research time from hours to seconds. By providing a centralised, intelligent interface for the entire document archive, DLV Advies has significantly improved its operational efficiency and service quality.
The ability to query our entire historical archive in natural language has fundamentally changed how our advisors work. We can now find specific precedents and project details that were previously buried in thousands of folders, all while ensuring every answer is backed by our own data.
Want results
like these?
Tell us your challenge and we'll show you how we'd solve it — with a clear scope, timeline, and fixed price.
Talk directly with our AI specialists