Purpose & Scope
This on-demand training guides participants through designing domain-specific AI assistants for agri-food applications by integrating graph databases with LLMs. Participants will learn a four-step process: setting up a graph database, linking it to a smart LLM, developing a responsive user interface, and applying it to smart farming use cases.
The course includes hands-on coding sessions with tools like Python, Streamlit, GitLab, and commercial or open-source LLM APIs. Real-world examples such as Pan-Café and Soilwise demonstrate practical implementation.
This training is ideal for technical teams supporting SMEs with data-driven advisory tools in agriculture and food systems.
If you want to get more information about the training, please get in touch with Marijke.hunninck@ilvo.vlaanderen.be
Learning objectives
By the end of the training, participants will be able to:
- Set up and populate a graph database with domain-specific content using Python
- Connect a commercial or open-source LLM to the graph database and query textual data
- Build a prototype user interface in Streamlit to interact with the AI system
- Explain the benefits and limitations of combining LLMs with knowledge graphs for smart farming tools
- Evaluate potential agri-food use cases for deploying AI assistants
Learning outcomes
Participants will:
- Gain practical experience building AI assistants for agri-food applications
- Understand integration of graph databases and LLMs
- Be able to prototype and test interactive AI solutions
- Assess the potential and limitations of AI-assisted decision-making tools in smart farming
Who should attend?
EDIH participants, technical teams supporting SMEs, AI developers, data scientists, and other professionals interested in AI-assisted advisory systems for agriculture and food.
AI implementation
Ethics
Other
Services
Software / AI model
Arable farming
Food processing
Horticulture
Livestock farming
