Shakthi Vel

AI/ML Engineer with hands-on experience designing, training, and deploying machine learning and deep learning models for classification, prediction, and anomaly detection use cases. B.E. in Artificial Intelligence and Data Science, Noorul Islam University (2026, 7.9 CGPA). Skilled in Python, TensorFlow, PyTorch, and scikit-learn for model development, cross-validation, and hyperparameter tuning, with experience building automated ML pipelines (data ingestion, preprocessing, feature engineering, training, evaluation) and deploying models via FastAPI and Docker.

Built and deployed a live fraud detection system using ensemble models (XGBoost, stacking/voting classifiers), achieving AUC ~1.00 on a 1,000-document dataset. Immediately available, open to relocating to Bengaluru.

AI/ML & Applied Data Science Intern at Curve Logics (2026-01 – 2026-06)

• Developed and fine-tuned ML models for game data analysis across 4 model variants using PyTorch and scikit-learn, applying cross-validation, hyperparameter tuning, and evaluation metric tracking to reach production-calibre performance.
• Built 3 automated end-to-end ML pipelines covering data ingestion, preprocessing, feature engineering, model training, and evaluation — deployed via FastAPI and containerised with Docker, cutting manual prep time by ~60%.
• Integrated Hugging Face Transformers for NLP-based feature extraction, optimising model inference logic for accuracy and throughput across production datasets.
• Delivered 4 Power BI dashboards translating model outputs into real-time business metrics, collaborated with senior engineers on solution design, and documented pipeline behaviour for scalability and reuse.

B.E. in Artificial Intelligence and Data Science – Noorul Islam University (2023-04 – 2026-04)