adrienne
@adrienne
Building an AI bot in public (cont.) For anyone following my journey of creating gmfc101, I’m starting to work on improving the quality of the bot’s replies. Right now, the bot isn’t getting enough good context from our transcripts to answer questions well. Here’s the current flow (chunk-based RAG): - Use semantic search to find the 5 most relevant conversations across all episodes. - Provide the 5 chunks (conversation snippets) to the LLM, along with metadata (episode details, YouTube URL, etc.). - Return the LLM’s response. Current issues: 1. The 5 random chunks lack coherence with each other. 2. The chunks aren’t detailed or complete enough to help the LLM generate accurate/useful answers—especially with voice transcripts, which are conversational and full of “ums,” “ahs,” and tangents. 3. The bot recommends YouTube links, but no one wants to watch an hour-long video to get an answer to their question. Going to try a new approach (cont...)
3 replies
2 recasts
17 reactions
adrienne
@adrienne
The new approach I'm going to try: - Simple RAG is not working, I'm going to try a multi-step process to improve the quality of the context I'm sending into the prompt - Will continue to use semantic search to find the most relevant episodes where we talked about the topic the user is asking about, but instead of supplying those chunks as-is to the LLM, I'll use the episode metadata to retrieve up the full transcript of those episodes to get better/richer/deeper content - I should also be able to have the bot provide timestamps when recommending videos to watch Wish me luck!
5 replies
0 recast
7 reactions
KMac🍌 ⏩
@kmacb.eth
💜💜💜👏 Please keep sharing. Heading down a similar path & your experience sharing is helpful. I was thinking of @alexpaden’s juicr as I read your cast. Are you using it for cast search to augment
2 replies
0 recast
1 reaction
shoni.eth
@alexpaden
it doesn’t exist yet, but also it won’t support custom data currently such as episode transcripts, just data native to fc makes sense to explore tho
0 reply
0 recast
0 reaction
adrienne
@adrienne
I will keep sharing. Public note taking helps me too It's a good forcing function to zoom out and take assessment of where it stands and what the next smallest iteration should be
0 reply
0 recast
0 reaction
