How to Talk to Tabular Data – A Chatbot Case Study with Raiffeisen Tech and the RBI Group

The controller of your personal data is Sages sp. z o.o., based in Warsaw. We process your data contained in the form for the purposes of executing or concluding a contract, settlements, accounting and financial reporting, compliance with other legal obligations, for the purposes indicated in the consents, to establish, pursue, and defend claims, and for statistical purposes. You have the right to access, rectify, erase, restrict, transfer, object to, and withdraw your consent to your personal data. You have the right to file a complaint with the President of the Personal Data Protection Office, and to withdraw your consent. More information can be found in the Privacy Policy.

Please be advised that the event will be recorded without visible images of the participants.

Large Language Models (LLMs) are playing an increasingly important role in business, but their practical use in corporate systems is still limited. The most commonly applied approach – Retrieval-Augmented Generation (RAG) – works very well with documents, yet in many companies the key data comes in the form of tables. How can we enable natural language queries for such data?In this presentation, I will discuss a project carried out jointly with Raiffeisen Tech for the Raiffeisen Bank International Group. I will showcase how we designed and implemented a chatbot that enables natural interaction with tabular data, and what techniques we used to ensure both precision and speed of responses.

What you’ll learn from this session:

Agenda