Sai Surya Prashanth Reddy Yakkanti

Generative AI Engineer at Allstate

Designed and deployed enterprise-grade Generative AI and LLM solutions (GPT, LLaMA, multimodal models) for insurance analytics, claims automation, fraud detection, and customer intelligence. Built large-scale Retrieval-Augmented Generation pipelines integrating structured policy data, claims records, and unstructured documents to enable faster and more accurate decision-making. Led prompt engineering frameworks and agent workflows to improve LLM accuracy and standardize outputs across business use cases. Implemented end-to-end MLOps and GenAIOps pipelines using Azure ML, Vertex AI, MLflow, Docker, Kubernetes, and Weights & Biases for scalable deployment and lifecycle management. Mentored engineers, led technical workshops, and collaborated with cross-functional teams to drive production-grade GenAI adoption.

Machine Learning Engineer at General Motors

Led end-to-end machine learning initiatives for automotive operations including demand forecasting, supply chain optimization, production scheduling, and quality defect prediction. Built scalable data pipelines integrating ERP systems, IoT sensor data, relational databases, APIs, and cloud platforms; performed feature engineering and preprocessing for model development. Developed predictive and classification models using Linear and Logistic Regression, Random Forest, XGBoost, LightGBM, CatBoost, neural networks, and time-series models (ARIMA/ARIMAX). Implemented scalable ML systems using Azure ML, Docker, Kubernetes, and microservices for real-time and batch inference. Delivered analytics dashboards in Tableau and Power BI and applied explainable AI techniques (SHAP, feature importance) to support decision-making in manufacturing and quality optimization workflows.

Data Scientist at McKesson

Led large-scale healthcare analytics projects processing multi-terabyte structured and unstructured datasets with Python, R, SQL, SAS, Spark, and Hadoop. Implemented end-to-end data science pipelines (ETL, feature engineering, EDA, predictive modeling), reducing data preparation time and accelerating insights for clinical and commercial teams. Developed models for classification, regression, clustering, and time-series forecasting using Scikit-learn, XGBoost, LightGBM, and R. Delivered executive dashboards via Tableau, Power BI, and Shiny, enabling data-driven decisions that improved clinical outcomes and reduced costs.