Apr '25 • General Discussion 💬
Help with Pinecone!
I'm using n8n to scrape content from a web page and store those chunks as vectors in a Pinecone vector store. That part works fine.
What I want to make sure of is:
- When the same web page is updated in the future, I want the existing vectors in Pinecone to be overwritten, not duplicated.
How can I best handle this in n8n?
- Do I need to generate my own hash-based ID per chunk?
- Will Pinecone automatically update a vector if the ID already exists?
- What's the best pattern to follow in this setup?
Any help, clarification, or sample flows would be greatly appreciated 🙏
6
3 comments
Help with Pinecone!
skool.com/ai-automation-society
Learn to get paid for AI solutions, regardless of your background.
Leaderboard (30-day)
1
🔥
+7638
2
+6360
3
🔥
+6001
4
🔥
+5478
5
🔥
+1947
Powered by