How is the vector database applied in your project?

Awadb_kk
2 replies
Hi, I hope you're having a pleasant day. I am an AI Vector Database Researcher, dedicated to discovering new technologies and methods that optimize the storage, retrieval, and manipulation of high-dimensional data vectors. I have been keeping a close eye on the development and achievements of your company. I am highly interested in your company's use of vector databases in your products, as I believe it holds significant importance in enhancing data processing and analytical efficiency. As a company at the forefront of the AI field, I would like to learn more about the specifics of how you integrate vector databases into your products. I am particularly interested in understanding the following aspects: How do you integrate vector databases into your products? Is it through in-house development or by adopting third-party solutions? In which application scenarios does your product utilize vector databases? How do they contribute in these scenarios? What key factors did you consider when selecting and designing vector databases? Such as performance, scalability, query speed, and more. Does the use of vector databases in your product help improve data similarity search and analytical capabilities? Are there any related success stories or customer feedback? Could you share some practical experiences and outcomes of applying vector databases in your product? I am genuinely interested in your company's experience and insights in this regard, and I would like to gain a deeper understanding of your approach. If possible, could I arrange a brief discussion with your technical team to further understand how vector databases are used in your product? Your time and insights are greatly valuable to me. Thank you for taking the time to read my message. I look forward to your response. Regardless of the outcome, I am excited about your company's achievements and anticipate the possibility of connecting with you in the future. Wishing you and your team all the best! Best regards!

Replies

Luke Skyward
Hey, I'm using Supabase in the project that I'm launching tomorrow (PodcastGPT). Supabase supports Postgres which has pgvector library that allows you to do vector search. It works perfectly fine for my use-case! The vector search allows me to find podcast sections that are most relevant to user questions. The results of the search are used as a part of RAG (Retrieval Augmented Generation) using Langchain!
Thomas Hansen
We built our own as an SQLite plugin based of FAISS. We're creating SQLite virtual/shadow tables, that stores the FAISS index as a binary blob object. You can check it out if you search for AINIRO ChatGPT Chatbot ...