Data Scientist

• Designed and deployed an end-to-end ML platform for trader behavioral segmentation and early-stage persona prediction using KMeans clustering (unsupervised) and Random Forest classification (supervised), serving predictions via REST API in production. • Engineered scalable data pipelines with SQL, Pandas, and PySpark to process trader-level behavioral features across millions of records; architected for large-scale dataset support. • Built production MLOps workflows using MLflow for experiment tracking, model versioning, and registry-based deployment; implemented automated validation, monitoring, and inference logging. • Containerized services with Docker and deployed on AWS using S3, ECR, and SageMaker endpoints for scalable model serving. • Leveraged Databricks and Spark DataFrames for distributed feature engineering and model experimentation. • Built FastAPI-based REST APIs and a Streamlit dashboard delivering segmentation insights and behavioral analytics to stakeholders.