Mandar Kadam

I’m a final-year B.E. (AI & ML) student who builds end-to-end machine learning systems across computer vision, fraud detection, and time-series forecasting. I work end-to-end across the ML lifecycle—data collection and preprocessing, model selection and evaluation, and deployment through practical interfaces like Streamlit and Flask—so prototypes become usable products.

In my Machine Learning & Software Engineering internship at Commonwealth Bank Pvt. Ltd., I analyzed large-scale operational datasets using ML pipeline standards, translating business objectives into structured, data-driven specifications. I also applied rule-based/heuristic classification logic to cybersecurity datasets to detect inconsistencies—communicating model-driven insights clearly to non-technical stakeholders.

My projects reflect my focus on measurable outcomes and production thinking: I built an image steganography & protection system with a full CV pipeline (dataset creation, annotation workflows, iterative evaluation, and Flask deployment). I developed an end-to-end payment fraud detection system with feature engineering and model selection under class imbalance, then deployed it via Streamlit with real-time scoring and explainability visuals; for cyclone prediction, I implemented time-series modeling, early anomaly-driven signatures, and a live monitoring dashboard for operational decision-making. I’m drawn to teams that value strong engineering discipline, thoughtful evaluation, and deploying AI where it matters.