[Integrated and lightweight photovoltaics]

Share this job offer

• Facing the challenges of our time
• The is offering a student project or internship to develop and validate an adaptive equipment-intelligence framework for semiconductor manufacturing using machine telemetry and high-frequency sensing
• Using semiconductor die bonding as the experimental platform, the project will investigate whether acoustic emission and other sensor signals can provide robust information about equipment and process state across changing recipes and operating conditions
• Machine-learning methods will be developed for:

Quality prediction

• Anomaly detection and virtual metrology, with emphasis on generalization across recipes and operating conditions
• Uncertainty-aware adaptation using limited data from new process conditions
• The project will quantify which sensing modalities and signal representations remain informative under process changes and determine how much new data is required to adapt the monitoring system to a new operating regime
• Design and instrument an experimental setup for semiconductor die-bonding measurements
• Acquire, synchronize, document, and preprocess machine telemetry, acoustic-emission data, and other high-frequency sensor signals
• Develop machine-learning approaches for quality prediction, anomaly detection, and virtual metrology
• Evaluate the robustness and generalization of sensing modalities and signal representations across recipes and operating conditions