KURUVA SURENDRA KUMAR

Built an end-to-end asset-management analytics pipeline: ingests multi-ticker market data (yFinance), engineers multi-asset risk matrices, and forecasts benchmark trends using Statsmodels AR models and XGBoost/Scikit-Learn.

Built an item-based collaborative filtering recommender on the MovieLens dataset using Scikit-Learn cosine similarity over a SciPy sparse (CSR) matrix, avoiding heavy Pandas matrix conversions to keep RAM usage low.

Built a statistical validation pipeline combining deterministic checks with embedding-based similarity search (Sentence-Transformers + FAISS) to detect data leakage.

A high-performance, real-time automated attendance tracking application. This system utilizes a lightweight computer vision pipeline to capture facial data through a webcam, match features against registered profiles, and seamlessly log entry records into a local database via a FastAPI backend.

ClipGenius is an intelligent, privacy-first Multimodal AI pipeline built to extract value from video content completely locally. The application splits video data into two parallel tracks—audio (linguistic processing) and visual (computational vision)—to generate a comprehensive text summary and detect visual highlights.

An end-to-end data analytics project that determines optimal product price points to maximize business revenue.