I am currently working an interest project supporting international student at my University as alumnus. The program, acts as a consultant, is assisting new-comers with up-to-date immigration policy, university brochure, and study permit/visa application, and many more features in development. The software currently utilizing RAG from Llama Index and LLMs Open AI Agents with Tools.
Currently, my only concern is how to scale the vector database for more accurate retrieval. I have not test the scaled, large vector database yet. I am still adding new features and function callings for a smooth pipeline. Other than that, my only concern still integrating a larger vector database. Any advice on this topic?
Please feel free to play around, I published the app on Streamlit share.
Additionally, feel free to add me on Linkedin as well. If you have any questions, direct message me, we can collab on this project for further development. This project means a lot to me because I am an international student myself and believe in this product would be a great support for new student to a different country.