UiPath shifts to shared GPU fleet on Google Cloud for agentic AI

Full analysis

UiPath, a market leader in enterprise agentic automation (software that deploys autonomous agents to reason, decide, and execute complex business processes), has re-architected its GPU infrastructure on Google Cloud to support large-scale intelligent document processing and agentic AI workloads. The company moved from isolated, on-demand GPU clusters to a shared fleet managed by its Machine Learning Services platform, using Google Cloud AI Hypercomputer with A3 VM instances (NVIDIA H100 GPUs) for training and G4 VM instances (NVIDIA RTX Pro 6000) for inference, scheduling capacity in advance via Google Cloud's Dynamic Workload Scheduler. The shift solved three core problems: spiky workloads that forced over-provisioning, H100 supply bottlenecks, and operational overhead from maintaining dedicated regional clusters. Early production results include Omega Healthcare automating over 100 million transactions at 99.5% accuracy and Thermo Fisher Scientific processing 53% of invoices without human involvement, cutting processing time by 70%.