The rocknbot repo contains code for a state-of-the-art RAG application. The application logic is implemented in a FastAPI program that can be accessed via Slack or the web. A Slack application and a web application have been provided to demonstrate usage.
Lil Lisa is an application that is responsible for integrating with Slack and handling incoming user events. It acts as a bridge between Slack's platform and the conversational AI experience provided by API calls to a concurrently-running FastAPI application.
LilLisa Web is a web application frontend for LilLisa Server.
Built using FastAPI, LilLisa Server is an AI question-and-answer program that enables users to ask natural language questions and receive insightful answers based on their personalized data.
LilLisa Server uses LanceDB as the vector database, providing blazingly fast retrieval times and LlamaIndex to handle complex queries.
LilLisa Server is able to grow its knowledge base and get better at answering user questions as more documentation is added and conversations are stored.
Zipped folder containing code that was used to evaluate the retrieval techniques.
PDF file documenting the results of the experimentation.
PDF file that goes into a deep-dive behind the experimentation and its results.