Behind the scenes of ChatGPT & Co. was the topic of the 6th Afterworkseminar
On 23rd May 2025, the Centre for Continuing Education and Knowledge Management (ZWW) at OTH Regensburg once again organized the sixth Afterworkseminar under the motto ‘Knowledge in the Evening’. The event was well-attended, with around 60 participants gathering to discuss the topic ‘Behind the scenes of ChatGPT & Co.’.
Prof Dr. Timo Baumann from the Faculty of Computer Science and Mathematics at OTH Regensburg, and Dr. Manuel Kirschner from BSH Haugeräte GmbH, were the keynote speakers for the evening, providing valuable insights into the world of AI and chatbots.
Modern chatbots demystified
Prof Dr. Baumann delved into the realm of artificial intelligence, focusing on the demystification of modern chatbots. He explained how modern chatbots are built on Large Language Models (LLM), tracing back their history to the first language models over a hundred years ago. Through technical advancements like neural networks and the processing of large datasets, chatbot technologies such as ChatGPT have seen significant progress.
One of the core elements discussed was the use of Large Language Models (LLM) along with Retrieval Augmented Generation (RAG) and Function Calling, enabling natural dialogues with chatbots. Prof Dr. Baumann also shared details about an ongoing project at OTH Regensburg where an AI chatbot is being developed to assist with student queries and access relevant information.
Generative artificial intelligence with Langflow at BSH
Dr. Kirschner shed light on the practical applications of AI at BSH, a leading household appliance manufacturer. He showcased how generative AI, particularly RAG flows, are utilized within the company. Various projects like ‘AIM’ and ‘RAISE’ were highlighted, demonstrating the incorporation of generative AI in internal knowledge systems and software development processes.
Dr. Kirschner also presented the ‘BAIA’ project, showcasing a voice dialogue system for household appliances, exemplified by a video demonstration of a voice-controlled washing machine. The session explored the development of an LLM- and RAG-based chatbot using ‘LangFlow’ software, emphasizing its user-friendly interface for internal knowledge sources.
The event spurred engaging discussions among the attendees, extending late into the evening during the subsequent networking session. The ZWW at OTH Regensburg has announced plans to integrate AI topics more extensively into their future education programs, including the introduction of a part-time university certificate in AI starting from September 2025. For further details, visit www.oth-regensburg.de/weiterbilden.
