Clockwork.io

Clockwork.io builds software-driven AI fabrics to increase GPU cluster utilization and keep AI workloads running through hardware and network failures. The company serves enterprises, neoclouds, research organizations, and AI infrastructure providers such as LinkedIn, Together AI, and NScale by providing cross-stack observability, workload fault tolerance, and performance acceleration that routes and paces traffic to avoid congestion. Its platform includes products like TorchPass for preserving training progress without code changes, LinkPass to handle link flaps, and FleetLens for automated cluster auditing and fault localization.

The open role at Clockwork.io indicates that the company is building a dedicated TorchPass Product Engineering team focused on making its infrastructure software installable, supportable, and operable by customers on their own clusters. The open position is a senior-level software engineering role located on site in Palo Alto, California. The team structure emphasizes experienced infrastructure software engineers who can own customer-facing surfaces, handle installations and configurations, work directly with clients during evaluations, and collaborate with core engineering teams to integrate with various schedulers and job launchers such as Kubernetes and Slurm.