"Builders Nepal team on AMD Developer Hackathon: ACT III Hackathon"

Team Idea

We want to build a KV cache optimization using simple alogrithm as we are genuinely interested in computer architecture and low level working of computer. We want to participate here because the main constraint in LLMs currently is KV cache. We want to contribute in that sector and we think the space given by the community will help us in researching that.

Team Members

Saurav Katwal

sauravktl1730759

AI Engineer

Bibek Gyawali

gyawalibibek0262697

Submission

LineGuard: Predictive Maintenance Copilot on AMD

LineGuard spots machines drifting toward failure about a day earlier than fixed alarms, predicts when they will trip, and an AI copilot on AMD GPUs explains why, cites the plant's manuals and drafts a work order: breakdowns become planned repairs.