Hi everyone, I've been lurking for a bit and finally decided to post. I'm in the middle of a classic demo fatigue spiral, and I'm hoping you can help me cut through it.
We're a team of about 50, and our infrastructure is a bit of a patchwork—some apps and data are on AWS, others are on-premise legacy systems. We need to implement a semantic search solution that can securely index and search across both environments. The dream is a unified search bar for our internal docs, support tickets, and code repos. My boss has given me a firm budget that basically rules out the huge enterprise platforms.
I've been looking at Relevance AI, and on the surface, it seems like it could fit. But I'm struggling to find clear info on how it handles hybrid setups. Is the agent framework overkill if we just need search right now? Also, I've seen a few mentions of Pinecone and Weaviate being more "infrastructure-native."
So my core questions are:
1. For a hybrid cloud/on-prem environment, is Relevance AI's architecture a good fit, or does it add complexity?
2. How does it compare, cost-wise and setup-wise, to rolling something with Weaviate or Elasticsearch for under 100 users?
3. Does anyone have real-world experience with the trial? I'm worried about hitting a wall after the POC when we scale to our full document set.
I'd be so grateful for any hands-on insights. The marketing sites all make it look easy, but I know the devil's in the details with these integrations.
New here!
Just my two cents.