Ph.D. candidate at MTSU researching federated learning, blockchain security, and privacy-preserving machine learning. Building systems that are both intelligent and trustworthy.
About
Highly motivated Ph.D. candidate in Computational and Data Science at Middle Tennessee State University with a strong foundation in Software Engineering. Passionate about leveraging machine learning, data analytics, and cloud technologies to solve complex real-world problems.
Experienced in developing and deploying ML models, conducting research in backdoor detection, privacy-preserving learning, and fake news detection. My work spans autonomous systems, healthcare AI, blockchain security, and large-scale data processing.
Before pursuing my Ph.D., I spent over four years as a Data Analyst at Urja Tech in Nepal, where I led data initiatives, architected SQL databases, and built Tableau dashboards that drove operational decisions.
Experience
Publications
Projects
Defense framework for distributed ML systems. Reduces attack success rates to below 15% through adversarial filtering and anomaly detection in federated environments.
Federated learning framework using satellite imagery for environmental monitoring. Achieves 0.85 IoU while keeping raw data local — no centralized data exposure.
Serverless AWS solution using BERT and LSTM models achieving 94.9% classification accuracy. Scales automatically via Lambda and API Gateway.
PySpark and Airflow batch processing pipeline for large-scale transaction fraud detection. Achieves 94.9% accuracy with automated scheduling and monitoring.
Generative adversarial networks applied to personalized educational content generation. Enhanced comprehension scores by 24% in pilot studies.
Ensemble of KNN, SVM, and XGBoost classifiers on chest-worn sensor data. Improved stress detection accuracy by 18% over baseline models.
Full-stack Next.js and Node.js SaaS platform for AI consulting services. Reduced client onboarding time by 35% through automated workflows.
Integrated Alteryx, Power BI, and SQL data pipeline for campaign performance analysis. Increased marketing ROI by 22% through data-driven targeting.
Skills
Talks