Master's Thesis (f/m/d) - Implementation of an OKF-based LLM Chatbot for Aerospace

Deutsches Zentrum für Luft- und Raumfahrt e.V.

ThesisGerman needed: not said

Still open on 1 Nov

You apply on the company's own website.

Key facts

Pay: Paid by federal public service collective agreement
Language needed: English
Subject you study: Computer Science, Computer Engineering, AI Engineering, Robotics

Overview of current state-of-the-art methods for using LLM technologies in specialized domains
Overview of robust evaluation methods for chatbot performance
Conducting a literature review on OKF and a comparison with RAG systems
Design and implementation of a state-of-the-art framework (OKF with or without RAG) optimized for both speed and accuracy
Introduction of a mechanism for evaluating the system regarding response time, precision and other possible metrics
Conducting an analysis of results comparing traditional RAG systems with the OKF approach using highly structured and hierarchical input documents
Development of an AI-supported compliance assistant that optimizes navigation through ECSS documentation and serves as an important and reliable reference work to accelerate decision-making for aerospace professionals

Enrolled in a Master's program (e.g. Computer Science, Computer Engineering, AI Engineering, Robotics) or a closely related field
Programming languages (e.g. C/C++, Python, Bash)
Knowledge in the area of LLM technologies
Knowledge of Git, unit tests and the basics of CI/CD as well as documentation
Ability to communicate in English, as supervision is conducted in English

DLR stands for diversity, appreciation and equality for all people. We promote independent work and the individual development of our employees in their personal and professional environment. For this purpose, our numerous training and further education opportunities are available to you. Equal opportunities are a particular concern for us; we therefore want to increase the proportion of women in science and management. Applications from severely disabled persons are given preference in case of professional suitability.

Overview of current state-of-the-art methods for using LLM technologies in specialized domains
Overview of robust evaluation methods for chatbot performance
Conducting a literature review on OKF and a comparison with RAG systems
Design and implementation of a state-of-the-art framework (OKF with or without RAG) optimized for both speed and accuracy
Introduction of a mechanism for evaluating the system regarding response time, precision and other possible metrics
Conducting an analysis of results comparing traditional RAG systems with the OKF approach using highly structured and hierarchical input documents
Development of an AI-supported compliance assistant that optimizes navigation through ECSS documentation and serves as an important and reliable reference work to accelerate decision-making for aerospace professionals

Enrolled in a Master's program (e.g. Computer Science, Computer Engineering, AI Engineering, Robotics) or a closely related field
Programming languages (e.g. C/C++, Python, Bash)
Knowledge in the area of LLM technologies
Knowledge of Git, unit tests and the basics of CI/CD as well as documentation
Ability to communicate in English, as supervision is conducted in English

DLR stands for diversity, appreciation and equality for all people. We promote independent work and the individual development of our employees in their personal and professional environment. For this purpose, our numerous training and further education opportunities are available to you. Equal opportunities are a particular concern for us; we therefore want to increase the proportion of women in science and management. Applications from severely disabled persons are given preference in case of professional suitability.