We know how hard it can be to ramp up and learn the ins and outs of a new company.
- “Who should I talk to about customer onboarding?”
- “What was that project the onboarding team shipped in June, that had a massive impact on step 3 completion rate?”
Instead of asking someone the same question that’s been asked hundreds of times before, it’s more efficient to find answers in existing documents and past conversations. The problem is, this data is spread out across dozens of workplace apps, with search features that all work differently. That’s why we’ve created Sleuth, an open source library that allows you to search through your company’s entire history using natural language. It understands the intent of your question, not just the keywords. Here’s a demo:
https://www.loom.com/share/71625cce862f4d4ea12b8a87ad94e407
You can fork our repo (https://github.com/getsleuth/Sleuth) and try it right now, or book a 15 min call (https://calendly.com/triton-founders/sleuth-feedback) with us to share your feedback.
How does it work?
Vector embeddings are generated for slack messages using OpenAI’s text-embedding-ada-002 model and stored in a Pinecone vector database for easy querying.
How is this different from Glean?
Glean is great, but we wanted to introduce a product that anyone can fork, use, and customize without ever talking to a sales team. Building in public makes for better products.
What integrations do you support?
Just Slack to start. What other integrations would you like to see?