TL;DR. Our ticket-routing eval scored 0.94 for five weeks. The number was manufactured. We had built...
Our few-shot examples came from the eval set. The 0.94 was fiction.
TL;DR. Our ticket-routing eval scored 0.94 for five weeks. The number was manufactured. We had built...
A retrieval pipeline built by hand in six files — the data structures behind each stage, the four bugs that cost me the most time, and the answer it made up that I left in the repo...
I built and open-sourced PacketVoyage—an Agent Skill & MCP server that turns boring traceroute...
Introduction For the past 10 days, I took part in the 10 Days of AI Voice Agents — #VoiceForBharat...