Karel Hovorka retweeted
My team finished in 3rd place at this weekend's Prague AI Hackathon
Project title: Multi-agent AGI 😀
We built a prototype of a multi-agent system with seven LLM agents connected in a hierarchical structure. Initially generalists, these agents adapt and specialize over time as they work on a shared task while receiving continual feedback from the user.
Benefits: in an ideal case, this system can solve any task, build a game, self-improve, etc.
Our motivation:
- One single agent has limitations in terms of context size, but collaborating with other agents and splitting the task into smaller, more manageable parts allows the team to overcome these limitations and achieve successful problem-solving
- Adaptability and ability to learn from user feedback
Karel Hovorka retweeted
Třetí místo jde GPT robotických pomocníkům, teda týmu, který je vytváři.
Posledni tym c.08 - Vytváření týmu GPT robotických pomocníků, kteří dokážou koordinovaně řešit komplexní úkoly, například objednání květin, nebo vytvoření aplikací na vytváření na stagy od @MarekRosa, @vynalezce, @thefillm, @gohanigo a dalsich
Karel Hovorka retweeted
My colleagues (Jan Feyereisl @thefillm and Karel Hovorka @vynalezce) and I are going to participate (compete?) at Prague AI Hackathon on April 21-22
We already have our topic:
Society of LLM-driven agents that recursively self-improve
It will be fun!
Wish us luck! @GoodAIdev
Follow @pragueaihack