Samyukta Reddy Vanga

Developed and deployed AI-powered claims review models using Python, Scikit-learn, and TensorFlow, improving claims processing accuracy by 28% and reducing manual review time by 40%. Implemented NLP pipelines with SpaCy, BERT, and LangChain to extract, summarize, and validate unstructured medical documents, ensuring HIPAA-compliant automation of claims adjudication. Built scalable ETL pipelines with Apache Spark and Kafka for processing high-volume healthcare claims data, enabling near real-time anomaly detection and fraud prevention. Led development of a Retrieval-Augmented Generation (RAG) framework with GPT-3 and vector databases for physician support, providing context-aware patient history insights. Deployed AI/ML models in production using Docker, Kubernetes, and TensorFlow Serving on AWS and Azure, ensuring scalability, fault tolerance, and 99.9% uptime. Collaborated with cross-functional teams (data engineers, product owners, compliance) in an Agile/Scrum environment.

Developed and deployed AI-powered claims review models using Python, Scikit-learn, and TensorFlow, improving claims processing accuracy by 28% and reducing manual review time by 40%. Implemented NLP pipelines with SpaCy, BERT, and LangChain to extract, summarize, and validate unstructured medical documents, enabling HIPAA-compliant automation of claims adjudication. Built scalable ETL pipelines with Apache Spark and Kafka for high-volume claims data, enabling near real-time anomaly detection and fraud prevention. Led development of a Retrieval-Augmented Generation (RAG) framework with GPT-3 and vector databases to provide context-aware patient history insights for physician decision support. Deployed models in production using Docker, Kubernetes, and TensorFlow Serving on AWS and Azure, ensuring scalability and 99.9% uptime. Collaborated with cross-functional teams in an Agile/Scrum environment, driving sprint deliverables and regulatory alignment.

Developed and deployed AI-powered claims review models using Python, Scikit-learn, and TensorFlow that improved claims processing accuracy by 28% and reduced manual review time by 40%. Implemented NLP pipelines using SpaCy, BERT, and LangChain to automate claims adjudication in a HIPAA-compliant manner. Built scalable ETL pipelines with Apache Spark and Kafka for real-time anomaly detection and fraud prevention. Led the development of a Retrieval-Augmented Generation framework with GPT-3 and vector databases to support physician decision-making by providing context-aware patient history insights. Deployed AI/ML models using Docker, Kubernetes, and TensorFlow Serving on AWS and Azure to ensure scalability, fault tolerance, and 99.9% uptime. Collaborated with cross-functional teams in Agile/Scrum to align on sprint deliverables and regulatory requirements.