# RoX AI Studio
> Cloud-native platform for evaluating AI models on real embedded hardware (Renesas R-Car SoCs)

## Overview
RoX AI Studio enables engineers to upload AI models, run inference workflows, and benchmark performance on real hardware using hardware-in-the-loop (HIL) execution.

## Capabilities
- Run AI model evaluation workflows
- Quantize and compile models for R-Car hardware
- Execute inference on real SoCs (HIL)
- Compare model performance metrics (latency, throughput, memory)
- Support Bring Your Own Model (ONNX)

## Typical Workflows
- Upload model -> trigger experiment -> retrieve metrics
- Compare two models on hardware
- Integrate benchmarking into CI/CD pipelines

## Authentication
- My Renesas login (user-facing)
- API access via token (for automation)

## Links
- Website: https://aistudio.renesas.com

## Developers
- Shashank Bangalore Lakshman
- Yao Chou
- Pengfei Yan