Joppe van der Schoot
AI Engineer & Entrepreneur
Profile
AI engineer and entrepreneur. I have worked with AI since 2023, after a bachelor's in Artificial Intelligence and a master's in Innovation & Entrepreneurship at Radboud University. I build AI that people actually use: data pipelines, AI agents, RAG and document AI that turns messy sources, from supermarket photos to handwritten forms, into reliable data.
I work with Google Cloud and Azure, and treat privacy and the GDPR as a design requirement from day one. With agentic coding I deliver fast and maintainably, and I run workshops on it for development teams. My business background helps me bridge users, business and technology.
Skills
Selected projects
Pricingscanner.com
Co-founder
AI-driven price and product comparison that extracts, recognises and compares product data from thousands of products.
- Built an AI pipeline that automatically collects, structures and normalises product data from diverse sources.
- Computer vision: recognising products and attributes from real, often poor-quality supermarket photos.
- Embeddings for taxonomy assignment (Google Product Taxonomy) and similarity search for comparable products.
- Live recipe pipeline: concepts → tool-using agent that maps ingredients to real products → SEO recipe with real nutrition data → images (batch on GCP) → social via API.
Document AI for a heritage institution
AI Engineer (freelance, via an AI agency)
Handwritten inspection forms turned into a reliable dataset, with a repeatable pipeline the client runs themselves.
- Read scanned, handwritten forms with a multimodal LLM into JSON following a fixed schema; normalisation in code (dates, numbers, codes via the legend).
- Measured quality without ground truth: a second independent read per field type (species and count 98.3% identical) and a review list linked to the scans.
- Repeatable and private by design: caching and prompt versioning, cost logging per call; staff names are never read or stored.
Beste.nl
Co-founder
From manual data entry to a fully automated AI process for digitising every retail leaflet in the Netherlands.
- Fully automated the manual digitisation process of 15+ people with an AI pipeline.
- Leaflet data is read, normalised and made comparable in real time.
Experience
Own company
Founder / Freelance AI Engineer
Freelance assignments and own clients, incl. on Azure: document AI for a heritage institution, ML pipeline and dashboard on HR data (Exire), education apps (TMI). Runs agentic coding workshops.
Project Manager & Software Developer
Build and direction of the AI projects Beste.nl and Pricingscanner; managing a developer and mentoring interns.
Startup
Co-founder
AI solutions that strengthen higher education rather than undermine it; pilots at Radboud University and universities of applied sciences.
Education
MSc Business Administration (Innovation & Entrepreneurship)
BSc Artificial Intelligence