At Vinayak IT Solutions, I built and deployed end-to-end machine learning applications under mentor guidance as part of my final-year B.Tech curriculum. I independently delivered two projects with minimal supervision, working across data preprocessing, exploratory analysis, and feature extraction.
For my main internship project, I implemented an online advertisement click prediction pipeline using Logistic Regression and Gradient Boosting, with ROC-AUC evaluation. I also built a Streamlit dashboard for real-time predictions and data visualizations.
I developed an AI-based distraction alert system using computer vision to detect focused, distracted, and phone-usage behaviour. It combined head-pose detection and phone bounding boxes with voice, sound, screenshot, and Telegram alerts.
I also designed an LSTM-based energy forecasting system for smart grids, engineering time-series features and building an interface to visualize consumption trends and renewable energy contributions. My projects have taken me from raw data through model development to Streamlit deployment.
