Tainan startup Aireco collaborates with NVIDIA to create an AI resource recycling ecosystem, injecting circular economy momentum into traditional factories.

As AI technology thrives, a group of passionate young people has emerged in Tainan, the hometown of Jen-Hsun Huang, dedicated to promoting artificial intelligence and sustainable environmental practices. They co-founded Aireco (宜科循環)) to redefine the economic value of resource recycling with innovative AI solutions, aspiring to create an AI-driven resource recycling ecosystem that injects new energy into traditional industries and transforms waste management into a high-value circular economy model.

Aireco was selected as one of the first selected start-up teams in Taiwan by the well-known Plug and Play accelerator in Silicon Valley this year, and exhibited a video commentary technology operation mode in NVIDIA Inception in Computex 2025, the following is a brief introduction to Aireco's core technology, corporate philosophy and NVIDIA's Isaac Sim application scenarios, detailed company introduction and technical explanation can be viewed on Aireco's official website.

Aireco Introduction

Aireco ( has partnered with NVIDIA to develop an artificial intelligence recycling system, using computer vision to replace human eye recognition, and utilizing robotic arms to replace manual labor in sorting and categorization. Through data analysis and hardware-software integration, it can accelerate operations in traditional factories, providing businesses with AI solutions for waste management, and enhancing efficiency.

Aireco Recycling Process

Aireco applies computer vision and robotic arms to waste sorting operations, effectively replacing manual visual inspection and manual sorting processes through artificial intelligence, big data analysis, and simulation training, thereby enhancing recycling efficiency and accuracy. Its system can assist businesses in integrating hardware and software, accelerating the automation of traditional recycling factories, and achieving the dual goals of intelligence and sustainable operation.

Aireco develops three core technology products to build an intelligent recycling system.

RECO Vista: Enhances waste identification and classification through actual production line image data and AI deep learning algorithms, improving the value of material reuse.

RECO Delta: Utilizing robotic arms for highly efficient and precise sorting operations, significantly enhancing the speed of recycling material processing and material recovery rate.

RECO Cognita: Based on data science, it integrates the Internet of Things and edge computing to assist traditional enterprises in building a complete AI management system, effectively monitoring production processes and operational efficiency.

How does Aireco's technology work?

In order to train high-performance AI models, Aireco uses the NVIDIA Omniverse platform to train AI models, first design a 3D model of a product, and change the shape of the products to make them look like waste, you can shoot the actual product through a professional camera or recreate a 3D model of a product, and then load them into the NVIDIA Isaac Sim simulation real training environment, using physics effects to make these 3D objects drop on the textured background. NVIDIA Replicator is then used to generate images and annotations of the scene, and these fixed numbers of 3D models are generated to generate a variety of different images for training AI to further effectively identify waste categories and treatment methods.

Isaac Sim is a technology for the NVIDIA Omniverse platform that provides highly scalable simulation capabilities, supporting data generation, software validation, and machine learning throughout the complete process, assisting developers in accelerating the development and deployment of AI robotic systems in a safe and controllable virtual environment.

What is NVIDIA Isaac Sim?

NVIDIA Isaac Sim is a reference application built on NVIDIA Omniverse that allows developers to simulate and test AI-driven robotic solutions in a physically-based virtual environment.

Isaac Sim is fully extensible, allowing developers to create custom simulators based on the Open Universal Scene Description, abbreviated as OpenUSD, or to integrate core Isaac Sim technology into existing testing and validation workflows.

The working principle of Isaac Sim

Isaac Sim operates three basic workflows: generating synthetic data for training and fine-tuning robotic base models, conducting software-in-the-loop testing on robotic stacks, and performing machine learning through Isaac Lab.

This article discusses the collaboration between Tainan startup Aireco and NVIDIA to create an AI resource recycling ecosystem, injecting circular economy momentum into traditional factories. It first appeared in Chain News ABMedia.