Working Student in Development of Retrieval-Augmented Generation, Pipeline Components (f/m/div.)

During your assignment, task alignments are possible including investigating custom parsing methods for technical documents, fine-tuning and extending LLMs for domain-specific use in industrial hydraulics, and evaluating prompt-tuning methods for tailored results from LLMs. The team mainly works remotely, with mobile working within Germany as the standard mode. Monthly visits to the common team on-site days are required. We're looking for someone enthusiastic about bleeding-edge technology willing to experiment with procedures and tools that have rarely been tried before.