Artificial Intelligence
Work description
The student will begin the work with a review of the state of the art on agentic AI systems, large language models, retrieval-augmented generation techniques, and retrieval and ranking methods applied to biobanks and precision oncology. Next, the student will analyse, integrate, and structure the heterogeneous data associated with PDX models, including clinical, molecular, pathological, genomic, imaging, and experimental information, in order to create a unified representation. The student will also be responsible for developing an agentic AI system capable of interpreting user requests expressed in natural language, coordinating specialised tools, and performing the retrieval, filtering, and re-ranking of PDX models according to complex selection criteria. The student will also be responsible for developing a conversational interface for interacting with the system and transparently presenting its recommendations. Finally, the student will validate the prototype using realistic use cases and biobank queries, assessing recommendation quality, information retrieval performance, usability, and the potential for reusing the solution in other biobank management and precision oncology contexts.
Academic Qualifications
Bachelor's degree in computer science, Artificial Intelligence, Data Science, or equivalent.
Minimum profile required
Proven knowledge and experience in the usage of Large Language Models API and tools (e.g. HuggingFace library).Advanced knowledge of English (primarily reading and writing).
Preference factors
Enrolled in a Msc in Artificial Intelligence or equivalent. Previous experience in agentic systems and AI in industry or research. Previous participation in research projects. Previous experience in writing and publishing scientific articles.
Application Period
Since 17 Sep 2026 to 02 Oct 2026
Centre
Artificial Intelligence and Decision Support