Data Scientist at Capgemini

Designed and deployed production-grade RAG systems using Azure OpenAI and vector databases, enabling domain-specific knowledge retrieval of 100k+ documents for enterprise users. Developed end-to-end ML pipelines including data preprocessing, feature engineering, and model training for classification and anomaly detection across enterprise datasets. Integrated ML/AI models into modern cloud architectures using Python, FastAPI, Docker, and Azure for scalable production integration. Established MLOps and versioning best practices to ensure model maintainability, monitoring readiness, and continuous data quality validation. Collaborated with architects and full-stack developers to ensure AI algorithms met functional, performance, and integration requirements. Authored detailed technical documentation covering RAG design choices, architecture decisions, and operational guidelines.