Aranya raises $11M to turn bare-metal servers into AI clusters in less than 48 hours

Aranya Inc., a startup that claims to be able to convert bare-metal servers into AI-ready GPU clusters in under 48 hours, has raised $11 million in funding to launch its operations. The company aims to tackle a major issue in the AI infrastructure market. Aranya was founded last year and has already managed over $500 million worth of GPU hardware for major AI inference providers and data centers.

Their main innovation is an open-source engine called clusterdOS, built on top of Kubernetes, which simplifies the deployment and maintenance of infrastructure through declarative configuration files. It automates containerized workload orchestration, networking, and storage resource management across different hardware environments.

ClusterdOS addresses Kubernetes' limitations by not only rescheduling a failed pod but also diagnosing the root cause, including GPU thermal issues, error-correcting code errors, and networking faults. It can manage heterogeneous GPU infrastructure, custom storage, and networking, transforming any hardware into a self-healing cluster within 48 hours.

Aranya's system uses agents that live inside the cluster and adapt to its needs, ensuring proactive issue detection and remediation. A natural-language interface allows engineers to manage infrastructure using simple commands, while strict permission controls prevent unauthorized access. The company also promises significant operational improvements, such as reduced downtime and optimized costs through proactive symptom detection and workload right-sizing. With the new funding, Aranya plans to grow its team and launch a full multicluster interface.