RunAnywhere

The default way of running on-device AI at Scale

Y CombinatorWinter 2026ActiveB2BInfrastructure

What they do

Problem

Deploying and managing on-device (edge) AI across many heterogeneous devices is complex due to differing hardware/runtime constraints, large models, and degraded performance under tight memory and power limits.

Solution

RunAnywhere provides an SDK to run AI models on-device plus a centralized control plane to manage deployments, enforce policies, and measure performance and outcomes across fleets of devices.

Who it's for

Enterprises and developers shipping edge AI applications across large fleets of devices.

Founders

Sanchit Monga

Co-founder & CEO

Former Intuit engineer building RunAnywhere, the infrastructure layer for deploying fast, private, multimodal AI on-device at scale. Deep background in mobile SDKs, platform tooling, and developer products, including systems used by 50M+ active users. Previously founded products across consumer discovery, context management, agentic documentation, and mobile testing, and now focused on making on-device AI production-ready across mobile, edge, and embedded devices.

LinkedIn · X

Shubham Malhotra

Founder

Co-founder & CTO of RunAnywhere (W26). Built MetalRT: the first complete multi-modal inference engine for Apple Silicon. Custom Metal GPU kernels that pushed on-device voice AI from 900ms to ~110ms. Ex-Amazon EC2 Spot ($100M+ ARR), Ex-Microsoft Azure. Peer-reviewed researcher.

LinkedIn · X