Harsika A S

I am an AI/ML Engineer with hands-on experience developing and optimizing machine learning and deep learning models for predictive analytics and classification. I focus on robust model development, feature engineering, and thorough performance evaluation using standard ML metrics.

My recent work includes building a quantum-assisted federated learning pipeline for tumor malignancy prediction, where I evaluated performance against classical federated learning baselines using accuracy, precision, and AUC. I have also developed a CNN-based web flood prediction system that achieved 92% accuracy and strong precision/recall/F1 performance.

In professional settings, I contributed as a backend/web development intern building and maintaining RESTful APIs with Node.js and MongoDB, assisted in backend deployment and monitoring, and optimized application performance for real-time use cases. I have experience integrating ML models into web/mobile applications and working with tools like TensorFlow, PyTorch, PennyLane, Flask, and OpenCV.

I bring a pragmatic approach to production ML: clear metrics, repeatable pipelines, and attention to scalability and deployment. I seek roles where I can apply advanced ML techniques, including federated and quantum-assisted methods, to build reliable, impactful solutions.