The project, named NutriGuard AI, aims to assess food quality and safety by combining several modern AI approaches. The application integrates image analysis based on computer vision models to detect defects or mold, simulated sensor data (pH, gases, temperature, storage time), and reasoning supported by language models. Together, these components enable a comprehensive and explainable assessment of food-related risks.
Beyond the technical outcome, this project illustrates the educational approach of the Digital Life Sciences track: teaching students how to design complete AI systems that combine data, models, and interfaces, rather than focusing solely on isolated algorithms. Students worked with tools and methods widely used in industry — including web applications, AI services, and model orchestration — within a framework tailored to bachelor-level education.
NutriGuard AI demonstrates students’ ability to rapidly mobilize key AI concepts and apply them to concrete challenges in the life sciences. Projects like this fully reflect the objective of the Starting with AI course: to provide future engineers with the foundations needed to understand, evaluate, and integrate AI within life sciences domains.