Built ML models and automation tools that reduced inspection time, improved fault detection, and predicted equipment failures across two advanced manufacturing companies serving aerospace, defense, and telecom.
Projects
ML for Equipment Failure Prediction (Jabil)
Developed ML model to predict equipment failures in optics manufacturing
Identified high-risk failure patterns before downtime occurred
Built automated Python alerts for 5-person engineering team
Shifted maintenance from reactive to predictive
Quality Control Automation (TTM)
Automated quality control workflows for printed circuit boards
Built real-time analytics tools for defect detection
~10% faster inspection cycles
~15% better fault detection accuracy
By The Numbers
10% reduction in PCB inspection time
15% improvement in fault detection accuracy
5-person engineering team optimized with automated alerts
2 companies serving aerospace, defense, automotive, and telecom sectors
$34B+ combined annual revenue across TTM and Jabil
Tech Stack
Languages: Python, SQL
ML/Data: Scikit-learn, Pandas, NumPy, data preprocessing, feature engineering