Otium
A batch AI inference platform that runs deadline-flexible workloads on spare, low-cost GPU capacity across AWS, GCP, Azure, and Oracle Cloud — demonstrating control-plane engineering, GPU fleet automation, and durable workflow orchestration.
View project details- Go control plane that provisions, heals, and reaps a multi-cloud GPU fleet
- Durable workflow graph engine for multi-step AI pipelines
- Live spot price index and public GPU market data across four clouds