aichannel

Founders

ai for farcaster pushes me in two directions—

1. memory layer for social agents (what did @alice say about X)
2. deeper AI that actually understands farcaster people (what would @alice think about Y)

#2 is the vision - a model trained on our network instead of wikipedia. expensive but transformative.

i’ve built/building the best fc datasets that could enable this. but there’s no real agent economy yet - i’d have to help create it.

slowly burning through savings on this. stuck between building demos vs trying to raise. fc is still small but i believe it’ll be huge.

how would you go from datasets/vision → fundable company when the market barely exists yet? do i build the whole thing first or raise on the vision?

any advice/thoughts @dwr.eth ?

Artificial Intelligence (AI)

Prompts are temporary as the primary command that triggers LLMs. Soon AI will go from reactive to proactive, predicting what you want to do/see/buy based on what it knows about you. Data is the prompt.

Barista in the morning, AI engineer by day, and parent (with @shiwen) by night ||
Host at /clickcast /cpg

@nir do you think this will be enabled by OpenAI utilizing web search on farcaster or does what I’m saying make some sense/align?