Naga Vamsi Reddy Satti

Research Intern (Machine Learning) at Teesside University

Built and trained deep learning models to classify medical images of Monkeypox using PyTorch and TensorFlow; achieved over 92% validation accuracy on 10k+ medical images. Designed preprocessing pipelines to reduce overfitting and improve generalization by 28%. Evaluated model performance using precision, recall, F1-score, ROC-AUC and confusion matrices to reduce false negatives by 32%. Integrated models into a Flask-based web application enabling real-time inference with a latency limit of 300ms for clinical end users. Contributed to end-to-end ML workflow tooling, reducing training and deployment time by 40%.