Sanjeev Kadam

• Architected end to end predictive analytics solutions leveraging XGBoost and LightGBM to optimize risk stratification models, improving clinical outcome forecasting accuracy by 18% using advanced feature engineering. • Designed scalable ETL workflows using Apache Airflow and Databricks, orchestrating structured and unstructured healthcare datasets into curated feature stores supporting real time inference pipelines. • Developed transformer based NLP models with BERT and spaCy for medical text classification and named entity extraction, enhancing automated claims adjudication efficiency by 22%. • Led advanced experimentation initiatives including A/B testing, uplift modeling, and causal inference analysis to quantify intervention effectiveness, driving data informed strategy decisions across population health management programs. • Implemented Retrieval Augmented Generation pipelines using LangChain, embeddings, and semantic search over enterprise knowledge repositories, accelerating contextual decision support across actuarial analytics teams. • Deployed containerized machine learning services on AWS EC2 using Docker and Kubernetes, establishing CI/CD driven model monitoring frameworks tracking drift, ROC AUC, and performance degradation.

• Architected enterprise grade machine learning pipelines integrating regression, clustering, and anomaly detection models to uncover revenue leakage drivers, boosting forecasting reliability by 21% across financial analytics systems. • Designed advanced deep neural network architectures using TensorFlow and PyTorch including LSTM and CNN frameworks to capture seasonality and temporal dependencies within high volume transactional datasets. • Orchestrated statistically rigorous experimentation frameworks leveraging cross validation, hyperparameter optimization, and dimensionality reduction techniques, elevating F1 Score performance while minimizing overfitting across production workloads. • Engineered high performance analytical data layers using PL SQL, PostgreSQL, and Amazon Redshift, optimizing complex query execution plans and reducing enterprise reporting latency by 35%. • Directed full lifecycle model governance including containerized deployment, automated retraining triggers, drift detection mechanisms, and cloud native monitoring using AWS Lambda and Kubernetes orchestration. • Spearheaded large scale feature engineering and model optimization initiatives across multimillion record datasets, improving model generalization capability by 17% while strengthening robustness through advanced validation and performance benchmarking methodologies.

Relevant Coursework: Database Management Systems | Software Engineering and Project Management | Data Science and Big Data Analytics | Business Analytics and Intelligence | Social Media Analytics