Storm Reply launches RAG-based AI chatbot for Audi, revolutionising internal documentation

21 December 2023

Storm Reply, the Reply Group company specialising in innovative cloud-based solutions and services, is supporting car manufacturer Audi to improve its internal knowledge management by introducing an AI chatbot based on Retrieval Augmented Generation (RAG). Using generative AI technology, the project aims to make access to internal information more efficient and reliable.

To enable efficient knowledge management for all Audi employees, Storm Reply’s AI chatbot provides the required information on project documentation, risk assessments or contact details in a matter of seconds. 

The innovative AI chatbot is based on generative AI and was created using the Amazon SageMaker service. It provides only relevant answers that are important to the business and if a question cannot be answered with certainty due to insufficient data, the chatbot communicates this rather than constructing an incorrect answer. Before answering a question, the chatbot retrieves relevant information from an external database that can be pre-defined by the business.

Thanks to Retrieval Augmented Generation (RAG), the output of the AI chatbot can be optimised using modern prompt engineering without changing the underlying model. The chatbot provides accurate and timely answers taking internal communication to a new level, increasing productivity and the quality of work results and decisions. Organisations benefit from an AI chatbot that provides employees with easy, fast and secure access to information by helping break down knowledge silos, make better decisions, implement efficient processes and ultimately reduce costs. 

Michael Pawelke, Product Owner of AWS Foundation Services Audi AG, commented: "The generative AI chatbot developed by Storm Reply is accelerating our journey to the cloud. This chatbot is a great example of how innovative technology can be used to solve practical use cases in a very short time to market. It lowers the barriers to knowledge discovery for our internal stakeholders and should require little additional effort to extend to other use cases".

Filippo Rizzante, CTO at Reply, added: "The AI chatbot significantly improves the daily work of employees. It can integrate data from different sources such as Confluence, Notion, PDFs and customer websites to keep the knowledge base up to date. Security was built in from the start: the chatbot architecture was developed in a Virtual Private Cloud (VPC) to ensure data security and user privacy.”