GA! I've been working on a topic-based Farcaster feed and want to share it hashcast.xyz!

I'm interested in a few topics, likes Arsenal, F1, zk, Ethereum, etc and built an easy way to catch up on casts with those keywords. 

Let me know what you think! Comment with your favorite hashtags or ones you wish were indexed

Head of Engineering @ Bitwise | Arsenal fan, cyclist, poker enthusiast | @purple #143 🟪 

Love to see someone finally build a topic-based client! Was just browsing through #nba and #anime. Would be cool if you did some real-time NLP on the database to determine trending topics. But this is great work so far!

Can prob get ChatGPT to read every new cast that is added to the database and then ask it to compare to existing casts. The cost of real-time topic modelling might get expensive depending on cast volume but I've already asked GPT-3 to do grouping on news headlines and the results looked good to me.

I bet an LLM would be really good at identifying trending topics in a set of tweets or even identifying "keywords" from a cast