At SOFTRON, I built and evaluated machine learning models for real-time business applications and contributed to AI-driven analytics solutions. I worked with Python, TensorFlow, and data pipelines to support model development and deployment.
I built a fake news classifier that achieved over 99% accuracy using Logistic Regression, SVM, and Random Forest. I applied NLP techniques, added interpretability with LIME, and deployed an interactive Streamlit app for real-time predictions.
I also developed an XGBoost model to forecast weekly store sales and integrated its predictions into a Streamlit dashboard. For customer lifetime value prediction, I built a regression model and used Power BI for customer segmentation and behavioural insights.

