At Nico Mac TCR Pvt Ltd, I improved prediction accuracy by 15%+ across production machine learning models through feature engineering, SMOTE, and hyperparameter tuning. I also deployed models as Flask REST APIs and Streamlit apps so business teams could access predictions directly.
At Wipro Limited, I developed Python automation scripts that reduced manual effort by 30%, and used Python and SQL for data extraction, transformation, and validation. In my projects, I built an end-to-end churn prediction model that reached 88%+ accuracy and an OpenAI-powered Q&A system.

