"AgentCrafters team on AMD AI League Hackathon"
Team Idea
A plug-and-play reasoning layer that wraps any open-source model and builds a graph of candidate reasoning steps, which it scores, expands and prunes. It refines its own answer over several iterations, so it gives better responses than the base model alone, with no fine-tuning needed.
Team Members
.png&w=256&q=75)
Rithik Sachdev
rithiks
Submission
Submission draft in progress