My major concern is that I wanted to use Thrive, but I saw that you have deprecated this. I tried reaching out to Thrive twice, but the response is not coming from them. I wanted to know that you guys are making something for vector database or you can recommend me something for vector database because I want to work on this because these things are required for my clients.

Since Trieve is deprecated, you can set up a custom knowledge base using a vector database like Pinecone. Here’s a quick Node.js example to get you started:

js
import { PineconeClient } from '@pinecone-database/pinecone';
import OpenAI from 'openai';

const pinecone = new PineconeClient();
const openai = new OpenAI();

app.post('/kb/search', async (req, res) => {
  const { message } = req.body;
  const latestQuery = getLatestUserMessage(message);
  const embedding = await openai.embeddings.create({
    model: 'text-embedding-ada-002',
    input: latestQuery
  });
  const index = pinecone.Index('knowledge-base');
  const searchResults = await index.query({
    vector: embedding.data[0].embedding,
    topK: 5,
    includeMetadata: true
  });
  // Format and return results as needed
});

This approach gives you full control and is production-ready for client projects ([see full guide](https://docs.vapi.ai/knowledge-base/custom-knowledge-base)). Source: - [Custom Knowledge Base Documentation](https://docs.vapi.ai/knowledge-base/custom-knowledge-base)

Hey! Since Trieve is deprecated, I’d recommend checking out this guide on migrating to a custom knowledge base: 👉 https://docs.vapi.ai/knowledge-base/migrating-from-trieve It walks through the recommended setup and alternatives so you can keep building without blockers. Let me know if you want help choosing a vector DB or wiring it up 👍