"RTHack team on AMD Developer Hackathon: ACT II Hackathon"

Team Idea

Building practical AI agents on AMD GPUs in the cloud, focused on automation, fast inference, and real-world impact.

Team Members

Christian Quintino De Luca

rthroadtoheaven12327

CEO Technical CTO

Submission

ZMLC+ Local-First Token Router v29-final

ZMLC+ separates deterministic/verifiable tasks from lossy ones: rules, verifiers, normalization, and optional local inference answer the safe slice with zero counted Fireworks tokens, while risky tasks route to allowed Fireworks models.