HIGHLY Scalable Vector Search Tutorial in 12 Mins!!!

🔥 Sign up for Astra vector database today. 🔥
Use your business email address and get from $1000- $3000 in free credits and consulting with your subscription

Learn how to build a fully scalable production-grade Q&A based on Vector DB in just 12 mins. We use
1. Astra DB (Cassandra backend)
2. Cassio
3. Langchain
4. OpenAI
as our Gen AI Tech Stack

🔗 Links 🔗

❤️ If you want to support the channel ❤️
Support here:

🧭 Follow me on 🧭

[titusfx]

🎯 Key Takeaways for quick navigation:

00:00 📚 Choose the right Vector database for scalable AI solutions, like Astra DB with Cassandra backend, for building a Q&A system.
03:00 🛠️ Set up Astra DB, upload secure bundles, and configure Cassandra clusters to prepare for Vector DB usage.
06:17 📊 Create embedding vectors and load text documents to enable Vector Search for answering questions.
09:19 🧠 Understand that the document you upload determines the system's knowledge, and adding more documents improves responses.
10:43 📈 Monitor and analyze Astra DB performance, including read and write requests, storage, and data transfer for scalability.

[santhoshkrishnan]

I am a great follower of you. appreciate your great effort

[coolsai]

thanks a lot for the awesome tutorial 🥳🎊🔥

[zkiyyeller]

Thank you very much for this Abdul. What part is using the AI model? When I include my document and it is vectorizing it? At any point is my document being shared?

[srikarkodakandla-zciy]

What is the system prompt that was given for this ? Will this have previous questions context ?

[sauravmohanty]

So are you saying the other Vector DBs in market are not fully scalable ?

[prasanosara]

Hi Abdul! thanks for the tutorial. Could you please do a video on Chromadb or any equivalent opensource DBs?

[shobhitbishop]

Is it any better than Milvus? We are using milvus and seems fine as a open source scalable solution

[anandsamuyelu]

Can you please suggest any models to fine tune on large raw dataset for text generation

[lonebadatel]

hey can you make a video on using langchain multi retriever with open source models like llama2 7B chat or mistral 7B

[jackmartin]

Really Nice video Abdul, quick query, my data to be embedded is in using Supabase (Postgres), can I use this to index my Postgres DB on a continuous basis and do some filtering, I need to use this as an API which returns some results which I pass to my Prompts, can you share some details on this or other potential solutions for this?

[RohanVector]

I want idea for final year project sir. Please help