Clockwork.io
Fault tolerance and performance software that keeps large GPU clusters doing useful work
About
Clockwork.io builds software that sits between AI infrastructure and workloads to keep distributed training and inference running through GPU, NIC, link and node failures, plus fleet monitoring and network acceleration for GPU clusters. It was founded in 2018 from Stanford research by Balaji Prabhakar, Yilong Geng and Deepak Merugu, with VMware co-founder Mendel Rosenblum as chief scientist, and is led by CEO Suresh Vasudevan. On October 5, 2026 it announced $31M in new funding co-led by Premji Invest, Wing Ventures and Seligman Ventures, bringing total funding to $73M, and launched TorchPass Snapshots. LinkedIn, Together AI and WhiteFiber are named users.
- Website
- clockwork.io
- Added via
- web
- Ownership
- unclaimed
Where to go
Products
- Clockwork FleetLens — Always-on visibility and health checks across GPU clusters and fabrics · Saas · Enterprise
- Clockwork TorchPass — Zero-restart fault tolerance for distributed AI training · Developer Tool · Enterprise
- Clockwork Workload Acceleration — Software-only network acceleration for AI training over Ethernet, InfiniBand and RoCE · Developer Tool · Enterprise