PhD Position in Natural Language Processing and Large Language Models Karlsruhe Institute of Technology in Germany

The Karlsruhe Institute of Technology (KIT) is seeking a PhD candidate in the field of natural language processing (NLP), with a focus on large language models (LLMs) and multimodal language processing. This opportunity is part of an EU-funded collaborative project aimed at advancing human-machine communication through state-of-the-art AI technologies. The research group, Artificial Intelligence for Language Technologies (AIALT) at the Institute of Anthropomatics and Robotics (IAR), specializes in developing language technologies that use AI to break down communication barriers and enable real-time translation and understanding between people.

The project covers a range of research areas, including large language models, multimodal models with a focus on speech language models, multilingual models, LLM agents, and LLM-based reasoning. The position offers the chance to collaborate with leading academic and industry teams in the field of speech processing and NLP.

Applicants should have or be close to completing a master's degree in computer science or a related technical discipline. Strong programming skills are essential, and previous experience in deep learning is considered a plus. High proficiency in English is required for effective communication with international partners.

The position is fully funded, with a competitive salary according to the TV-L E13 public sector agreement, additional pension benefits, and public transport allowance. The successful candidate will have access to excellent computing infrastructure, opportunities to attend international conferences, and a wide range of training options. The group values diversity and encourages applications from women and individuals with disabilities.

To apply, candidates should send a cover letter and CV (including contact information for two references) to Prof. Dr. Jan Niehues at [email protected] by March 15, 2026. The starting date is flexible in 2026.