At O Clock Software, I developed reusable Python workflows for data processing, preprocessing, validation, and automation. I also wrote scripts for recurring operations and structured data preparation.
I built ChestXpert, a multi-label chest X-ray classification pipeline using ResNet-18 transfer learning on a 2,000-image development subset with 14 abnormality labels. I implemented preprocessing, model training, evaluation, Grad-CAM explanations, and a Flask REST API with a Streamlit review interface.
For FaceFare, I engineered a facial-recognition fare system that reduced simulated passenger boarding time from 6 seconds to 4 seconds per passenger. I built its Flask backend and implemented SQLite wallet transactions, face enrollment, and real-time dashboard updates.
I also built SalesPilot AI, a sales automation platform with explainable lead scoring and pipeline forecasting, and developed a vehicle-refueling safety pipeline combining four model inference stages. I’m a co-inventor on an Indian patent application for the FaceFare public transport payment system.
