Sameer User

I've built production ML pipelines and modular software services as a Data Science Intern at Celebal Technologies, processing 100K+ row enterprise datasets and reducing manual preprocessing effort by 60%.

I develop classification and regression models using Scikit-Learn, Random Forest, XGBoost, GridSearchCV, and k-fold cross-validation. My work achieved F1 scores of at least 0.88 on imbalanced datasets, while engineered predictive features improved RMSE by 18%.

I've also built a hybrid semantic search engine over 200K+ Wikipedia articles, combining BM25, Sentence-BERT, and FAISS with sub-100ms FastAPI queries.

My projects include an AI-powered resume matcher that reached 89% matching accuracy across 100K+ resumes in sub-50ms, plus an automated ML framework with preprocessing, ensemble modeling, Optuna tuning, and Docker deployment. I'm completing a B.Tech in Computer Science & Engineering (Artificial Intelligence) and enjoy applying strong Python and data structures fundamentals to practical AI products.