A LIVE SYSTEMS DEMONSTRATION · 2026
FROM MODEL
TO PRODUCTION
A live exploration of the systems
behind AI inference.
01 / 05
/chat-1
Direct inference
One model. One serving path.
02 / 05
/chat-2
Distributed inference
Same model. Different infrastructure.
03 / 05
/gpu
GPU telemetry
Watch the hardware respond.
04 / 05
/models
ML model serving
Traditional ML inference, served properly.
05 / 05
/terminal
Live terminal
Infrastructure, in motion.