Yisz

3 ms·

1. 1.▲
   by Yisz 2y ago

   How does it compare with the built-in human-in-the-loop feature from langgraph? Or CrewAI allows humaninput as well right?
2. 2.▲
   by Yisz 2y ago

   Very cool. Does this also generate the transcripts or only add search for videos with existing transcripts?
3. 3.▲
   by Yisz 3y ago

   We have a number of production RAG datasets (human-verified) and have custom generated evaluation datasets for a variety of companies. Feel free to reach out if you are interested: yi@relari.ai
4. 4.▲
   by Yisz 3y ago

   Very interesting comparison of RAG to traditional search development. I'm curious how large an evaluation dataset is usually necessary to compare results from different experiments of Google search?