Developing supervised machine learning models for detecting a critical reliability defect (electrical short), integrating it with SPC and interactive dashboards for rapid response.

Developed a RAG-based chatbot using open-source LLMs (LLama) to interact with quality records (NCR, MRB, 8D, etc.), streamlining information retrieval and improving accessibility.

Automated QA teams's 3 daily tasks (workflow) with custom-built web apps (Python, Flask, JavaScript, React) cutting manual work by 8+ engineering hrs/week.

Engineering interactive Tableau dashboards that transformed KPI monitoring from manual reporting to real-time analytics, enabling data-driven decisions, and reclaiming 4 engineering hours weekly.

Implementing unsupervised (clustering) anomaly detection on semiconductor manufacturing parametric data to identify outlier wafers.

Solved the company’s biggest customer issue—battery leaks—by implementing a two-factor Design of Experiments (DOE) to validate root causes and developed effective PCAs collaborating with cross-functional groups, reducing the battery leak failure rate to 0%.

Trained an ML classification model to automatically classify (~ 80% accuracy) operator-entered MES defects into 4M (Man, Method, Machine, Material) categories providing real-time analytics.

Developed 4 top-of-class Tableau dashboards to monitor key quality metrics - Torque, NCR, Battery Test, Cpk and provide real time email alerts.

Led APQP activities (Product Quality Plan) from scratch to finish by deploying PFDs, PFMEAs, control plans, SOPs, records of material, qualified lab documentation, MSA studies, packaging specifications, and initial process capability studies.

Validated ultrasonic weld inspection tool for assembly line use through linear regression and Gauge R&R, statistically comparing it against destructive testing to confirm its efficacy, thereby reducing inspection time and cost by 90%.