# RooflineAI

RooflineAI builds a retargetable compiler SDK that lets hardware vendors and embedded software teams deploy and optimize AI/ML models on edge devices (NPUs, MCUs, heterogeneous SoCs) without manual tuning. Built on open-source MLIR/IREE with proprietary hardware-specific optimization on top.

## Solutions
https://www.roofline.ai/
https://www.roofline.ai/ml-solution
https://www.roofline.ai/hardware-solution

## Products
https://www.roofline.ai/product-sdk
https://www.roofline.ai/product-runtime
https://www.roofline.ai/product-performance-dashboard

## Case studies
https://www.roofline.ai/case-studies/asynchronous-heterogeneous-execution
https://www.roofline.ai/case-studies/nxp-neutron-llm-enablement
https://www.roofline.ai/case-studies/dynamic-shapes-llms