AI reporting assistant
How a built-in (generative) AI assistant can support the information need in an underwriting process
Project objective
Allianz Trade specialises in trade-related insurance solutions that help businesses protect their financial stability. They primarily offer trade credit insurance, which safeguards companies against losses when customers fail to pay invoices.
For their commercial underwriting process, in which all related risks are assessed, Allianz Trade makes use of a global application to monitor and underwrite those risks.
The objective of this project was to facilitate users in the need for statistical information by integrating a GenAI solution into the application. The challenge lies in the fact that the data needed for effective decision-making is available, but is scattered in databases and documents. In addition, every request requires a different set of data to make a decision. Allianz Trade wanted an intuitive way to ask any type of context-related question, that is easily accessible.
Our solution
Finaps introduced a chat-window within the application, supporting the underwriting process. This chat-window is a gateway to interact with the Large-Language Model (LLM) deployed within the infrastructure of our client. With each conversation, the context for the underwriting process is provided to the LLM. The LLM is used to determine what data the application needs to retrieve using techniques like Model Context Protocol (MCP) and Retrieval-Augmented Generation (RAG). In addition, the LLM is instructed to generate an SQL-statement to retrieve data from the application’s own database. The data from the different sources is combined, and the LLM is used to produce a reliable and human-readable answer to the question asked.
“The LLM is used to determine what data the application needs to retrieve using techniques like Model Context Protocol (MCP) and Retrieval-Augmented Generation (RAG).”
“The LLM is instructed to generate an SQL-statement to retrieve data from the application’s own database. “
Industry
Technology
Mendix, Java, Generative AI
