Chandana C.

Designed and deployed complete Machine Learning (ML) pipelines using AWS SageMaker for real-time fraud and anomaly detection, processing millions of transactions with sub-second risk scoring latency. Engineered a customer churn prediction model via Scikit-learn and XGBoost, which involved exploratory data analysis (EDA), cross-validation, and feature engineering employing Pandas and NumPy. Developed LLM-powered Natural Language Processing (NLP) applications with Hugging Face Transformers and LangChain (RAG architecture), leading to a reduction in manual customer insight analysis. Architected MLOps pipelines utilizing Docker, Kubernetes, and MLflow CI/CD automation, significantly decreasing model deployment time and standardizing the model lifecycle across teams. Created real-time inference REST APIs using Flask, promoting seamless integration of production model scoring with enterprise systems. Conducted extensive feature engineering and data wrangling through Apache Spark and Airflow, converting large datasets into high-quality, model-ready data. Established frameworks for model monitoring, statistical analysis, and drift detection to maintain model performance and automate retraining in production. Developed executive Power BI dashboards that translated ML insights into actionable KPIs, facilitating alignment between data science outputs and business strategies. Ensured Responsible AI compliance across deployed models by implementing bias detection and model interpretability measures.

Analyzed extensive healthcare datasets, consisting of millions of patient records, using Python (Pandas, NumPy) and SQL, focusing on statistical analysis and data wrangling to identify clinical and operational trends. Developed classification and time-series forecasting models with Scikit-learn, which involved cross-validation and model evaluation within Jupyter Notebook, enhancing prediction accuracy for key healthcare metrics. Designed and refined ETL data pipelines to decrease data processing latency, thereby increasing data availability for downstream analytics teams. Created interactive dashboards in Power BI and Tableau for business intelligence reporting, resulting in improved reporting efficiency and reduced manual reporting cycles. Performed exploratory data analysis (EDA) and statistical hypothesis testing, leading to actionable insights that positively impacted healthcare operational workflows. Collaborated with clinical, engineering, and product teams to convert complex business requirements into scalable and audit-compliant analytics solutions. Maintained strict data governance and quality standards to ensure HIPAA compliance and regulatory alignment for all analytical outputs.