Venu AI

Project 1: I built and deployed a LightGBM regression model in Python to predict the lowest cost LED component configurations for customer lighting solutions. By training the model on historical pricing and technical data, I achieved 99.7% prediction accuracy and reduced cost estimation time from three days to under thirty seconds. This enabled faster customer responses and supported more competitive, data driven pricing decisions. Project 2: I automated the formatting and validation of over 100 LED driver specification sheets using Excel VBA to standardize inputs and eliminate manual errors. This automation improved analyst accuracy and efficiency by ~85% while ensuring consistency across customer facing technical documentation. The project demonstrated how lightweight internal tools can meaningfully improve operational workflows. Project 3: I diagnosed and analyzed over 20 inverter failures by conducting hands on PCB testing, including MOSFET analysis, thermocouple measurements, and current flow validation. By identifying root causes and failure patterns, I helped improve system reliability and reduce downstream defects for end users. This work strengthened my ability to combine hardware troubleshooting with data driven engineering analysis.