Sample: Dozer + LLM + Vector database + Langchain sample

A few days ago we published an article (https://getdozer.io/blog/llm-chatbot) describing how Dozer could improve hyper-personalization when used together with LLMs, Vector databases, and Langchain.

The article describes a hypothetical bank implementing an LLM-based chatbot and leveraging Dozer to create a unified customer profile and later on passing it to an LLM as a context to hyper-personalize the chatbot.

Based on this article, a complete working sample must be produced. Dozer should be configured to source from multiple datasets (like customer profiles, transactions, etc) and Dozer APIs to be integrated with langchain. A similar use case (as described in the article) of credit card products should be built.

Start by reading the linked LLM chatbot article to understand the proposed bank and credit-card use case. Then inspect Dozer's data-source configuration and APIs alongside the Langchain integration requirements; done means a complete working sample that combines customer and transaction datasets, builds a unified profile, and uses it as chatbot context.

Written by the indexing model from the issue text.