Two alumni of Greenwood High International School have secured first place in the poster competition at Purdue University's AI Research Showcase 2026 for their work on an artificial intelligence-based solution for semiconductor defect detection.
Rishikesh Madhuvairy (Class of 2025), now a freshman at Rice University, and Rahul Prabhu (Class of 2024), a sophomore at Purdue University, were recognised for their research project that explored the use of artificial intelligence to improve quality control in semiconductor manufacturing.
Their project focused on developing an AI-powered model capable of identifying microscopic defects in semiconductor wafers with greater speed and accuracy. The researchers said the approach could help improve chip quality, manufacturing efficiency and the reliability of semiconductor devices.
The paper, titled "Fusing Handcrafted Spatial Descriptors with a Lightweight CNN for Semiconductor Wafer Map Detection Classification," examined how defects in two-dimensional semiconductor wafers can be identified and classified based on where they originate during the manufacturing process.
"The premise for our entire project was to create a way to automate the prediction and classification of these defects that occur on a regular basis," said Rishikesh Madhuvairy. "Our model could further be integrated into a yield intelligence platform that can tell quality control engineers where and how, spatially speaking, defects are arising in silicon wafers."
Rahul Prabhu said, "Combining statistics with computer vision models is not new. But applying it to semiconductor defect classification with this kind of industry dataset is."
According to the students, the model uses eight spatial descriptors to identify the location and distribution of defects before processing the information through a convolutional neural network. They reported that the approach achieved nearly 13 per cent higher accuracy than a baseline computer vision model, while also being suitable for on-site deployment.
Reflecting on the collaborative nature of the project, Madhuvairy said, "I came to this project not knowing much about the computer vision aspect. I brought the theoretical knowledge of spatial classification, but to apply it in a commercial use model, then you need skills to develop the vision model. Rahul's expertise really came through as he took the reins of that part of the project, and I learned a lot from him."
Congratulating the alumni, Niru Agarwal, Managing Trustee, Greenwood High International School, said, "Congratulations to Rishikesh Madhuvairy and Rahul Prabhu on this outstanding achievement and for making Greenwood High proud on the international stage."
The students said they intend to further develop the model by increasing its capabilities, integrating it into a physical setup and exploring potential commercial applications. Madhuvairy is also pursuing projects in areas including renewable energy and wearable medical devices.
