Internship title: Integration of the Retrieval-Augmented Generation process in a Large Language Model for Sentinel-3 operational documentation. EUMETSAT is exploring the possibility to use Large Language Models (LLMs) integrated with Retrieval-Augmented Generation (RAG) to assist and enhance the capabilities to look up, query the operational and technical documentation baseline via the creation of an AI “virtual assistant” for the spacecraft operations team.

The intern will first develop an understanding of the team’s needs to balance efficient search and retrieval of large documentation sets with the need to keep EUMETSAT’s data private and protected, and how these needs can be met. They will also learn how to manipulate the documentation in a format which can be used by the RAG technology.

Finally, the intern will support the development of a virtual assistant by building on the work done by the EUMETSAT EDGA project and create a draft of document writing guidelines applicable to future missions to ensure documentation sets are AI-readable.

• The internships are open to bachelor and master students with mandatory internship requirements in relevant disciplines such as Computer science and spacecraft engineering.
• The official languages of EUMETSAT are English and French. It is necessary to be able to work effectively in English.
• Have the ability to work effectively in English;
• Be computer literate;
• Be intrinsically motivated and curious about the internship subject;
• Be able to work independently and collaboratively;
• Have the ability to take the initiative in researching ideas;
• Have the ability to collect, collate, conceptualize and present information clearly.

Additional specific requirements:

• Proficiency in Python;
• Software development and architecture knowledge;
• Knowledge of AI/Machine Learning methods.