AnkushKumar User

I’m a full-stack developer and ML practitioner pursuing my M.Sc. in Computer Science at NIT Trichy. I enjoy building production-grade web applications using the MERN stack and pairing them with end-to-end machine learning pipelines that solve real-world problems.

As a Machine Learning Intern, I built an ML-based student performance prediction system using academic, attendance, and demographic data. I achieved the best results with Random Forest among five compared classifiers, applying preprocessing, feature engineering, and class-imbalance handling with precision/recall and confusion-matrix evaluation.

In my projects, I focus on shipping complete, testable products. With JamX, I built a fully functional ed-tech platform with JWT & Bcrypt authentication, dual role-based views, Redux global state, and a RESTful backend in Node.js/Express.js/MongoDB—plus Postman testing for end-to-end API reliability.

I also love multimodal AI that’s practical for users. MultiQuery AI is a multimodal RAG system for PDF, image, and audio querying using SentenceTransformers, CLIP-style image embeddings, Whisper transcription, ChromaDB semantic search, and a context-restricted response pipeline—deployed via Streamlit.