Modern nanofabrication relies on complex tools, where small process deviations can lead to significant yield loss and wasted time. Despite this, current workflows depend heavily on manual logging and experience-based troubleshooting, limiting process visibility and making root-cause analysis slow and inconsistent. To address this, we propose a data-driven framework for structured troubleshooting and process monitoring. We collect user-reported data on failure modes, frequency, and time loss to identify high-impact issues across nanofabrication tools. Then, we evaluate structured validation steps as practical interventions to reduce failures. By replacing trial-and-error debugging with systematic, data-informed workflows, this work aims to improve diagnostic efficiency, reduce downtime, and enable more scalable and reliable nanofabrication processes.

I chose to participate in this SuperUROP because it is interdisciplinary, combining my interests in AI and semiconductor technologies. I am interested in exploring how data-driven approaches can improve hardware reliability in nanofabrication. I hope to deepen my understanding of areas such as anomaly detection and system-level diagnostics, while gaining hands-on experience working with real-world process data. My goal is to contribute meaningfully to our group’s efforts to improve tool reliability and develop more systematic, efficient approaches to troubleshooting in nanofabrication environments.