2/ Plan Prism - 🥉3rd place AI agents that take PRDs and automatically improve them via multi-agent collaborative debate m AI agents to replace PMs ankit tandon Sunanda Saxena

OpenAI’s biggest rival is shaking things up. Anthropic invited 200+ elite hackers to their SF headquarters to see what’s possible with Claude Here’s what we saw at the @AnthropicAI x @MenloVentures Builder Day Hackathon (🧵):

1/ China town mafia - 🥈2nd place Next gen anti-captcha that detects when AI agents powered by Claude’s computer use try to solve verification puzzles Honeypot for AI agents @_weichunnn @benxneo

3/ Robot use - 🥇1st place Combining Claude computer use with a robot arm and have the LLM perform real world actions Amazon robot controlled just by uploading the instruction manual to Claude

4/ Yap to App Controlling a computer only using voice and claude tools Combining Apple scripting with speech to text models to pilot your Mac @louzell_

5/ RAGs to Riches Text messaging agent to answer questions about CalFresh benefits (aka California food stamps) AI for welfare

Thanks again to @alexalbert__ @SoheilK @deedydas @CCgong @DarioAmodei for hosting + judging Still want to hack with Claude? @AgentOpsAI is hosting another hackathon with @AnthropicAI this weekend. RSVP below and follow @AlexReibman for the latest

@ankittandon1 @_sunandasaxena Why does everyone think the PM's job is just about writing a PRD? 🤔

@ankittandon1 @_sunandasaxena How effective was the PRD hack ? Plan prism - idea is promising - what about quality of output?

@ankittandon1 @_sunandasaxena @shravantickoo23

@ankittandon1 @_sunandasaxena But (as a developer) will you be able to blame it for your failures?

@ankittandon1 @_sunandasaxena This is a kid's play. I did this for a client with OAI api like 8-9 months ago. This is now a prompt not a app. The UI was a tough cause client wanted too many things, and that will result in too many round trips to API.

@ankittandon1 @_sunandasaxena @mreflow