SkyPilot
SkyPilot provides a control plane that unifies AI compute across Kubernetes, Slurm, and 20+ cloud providers into a single resource pool. The platform supports interactive development, batch jobs, online reinforcement learning, and production inference endpoints, with top deployments exceeding 1,000 nodes and 10,000 GPUs. SkyPilot also offers GPU Compass, a dashboard for comparing GPU offerings across 2,000+ instances.
- Artificial Intelligence
- Developer Tools
- Software Only
- AI Agents
Funding
Founders
Founder details are not available yet.
Product
Problem
AI teams face significant compute fragmentation, managing workloads across multiple cloud providers, on-premises clusters, and different schedulers like Kubernetes and Slurm. This fragmentation forces teams to navigate complex, provider-specific setups, leading to inefficient resource utilization, delayed development cycles, and high operational overhead.
Solution
SkyPilot provides a control plane that abstracts all AI compute—across neoclouds, hyperscalers, Kubernetes, and Slurm—into a single, unified pool of resources. This "AI supercomputer" allows teams to centrally manage and efficiently use their infrastructure, supporting interactive development, batch jobs, and online reinforcement learning. It also offers production-ready inference endpoints and GPU Compass, a dashboard for browsing and comparing GPU offerings across 20+ clouds and 2,000+ instances, all while keeping code and data within the customer's own cloud environment. SkyPilot started as an open-source project at UC Berkeley's Sky Computing Lab and is now used by hundreds of leading companies, with top deployments surging past 1,000+ nodes and 10,000+ GPUs.
Target Audience
Primary customers are frontier AI labs, enterprises, and research organizations running large-scale training, batch inference, or online RL workloads, including companies like Shopify, H Company, Nubank, and Covariant.
Features
- Unified control plane for managing compute across Kubernetes, Slurm, and 20+ cloud providers
- Native support for interactive development, batch jobs, and online reinforcement learning workloads
- Gang scheduling and multi-node job orchestration for distributed training
- GPU Compass dashboard for browsing, comparing pricing, and launching GPU instances across 2,000+ offerings
- Production-ready inference endpoints with one-YAML deployment, at up to one-tenth the cost of hosted alternatives
- SkyPilot Sandboxes for running untrusted, LLM-generated code on Kubernetes clusters, sustaining 50,000+ sandboxes per cluster with sub-second launches
- Agent-native capabilities with official SkyPilot Skill for launching workloads and managing infrastructure
- Enterprise governance features including SSO, RBAC, admin policies, secrets management, and audit logs