Deep learning practitioner focused on real-world medical AI and predictive modeling. I build models that don't just perform — they explain themselves.
I'm Sameer Singh, a machine learning engineer with a passion for solving problems that matter. My work sits at the intersection of deep learning, computer vision, and real-world impact — particularly in healthcare and predictive analytics.
I take models from a rough baseline to production-grade systems, focusing on what metrics actually matter — not just accuracy, but recall, interpretability, and clinical utility.
View GitHub →Deep learning model detecting pneumonia in chest X-rays using ResNet-50 with transfer learning. Boosted accuracy from a 66% baseline to 90.1% through fine-tuning, augmentation, and differential learning rates. Includes Grad-CAM heatmaps for model interpretability and a Gradio web interface.
Machine learning pipeline for predicting loan approval outcomes. Built with data preprocessing, feature engineering, and classical ML algorithms in Jupyter Notebooks — a solid foundation in end-to-end predictive modeling workflows.
Currently building new projects in medical imaging, NLP, and production ML systems.
Follow on GitHubOpen to collaborations, research projects, and opportunities in ML engineering. If you're building something intelligent, let's talk.