Vamshi Shetty

Designed and built Retrieval-Augmented Generation (RAG) solutions using LangChain, OpenAI APIs, embeddings, and vector databases to help users retrieve information from enterprise knowledge sources via natural language queries. Developed AI-powered document processing capabilities for search, summarization, and exploration across structured and unstructured documents. Implemented semantic search using embeddings and vector databases (including Pinecone) to improve relevance and contextual retrieval. Integrated enterprise data from SharePoint, relational databases, JSON, PDF documents, and web sources into retrieval pipelines. Built chunking, embedding generation, and indexing strategies for scalable enterprise search. Developed FastAPI-based REST APIs to integrate Generative AI features with enterprise applications. Deployed and maintained production services using Docker, Kubernetes, and CI/CD while supporting monitoring and issue resolution.

Partnered with business and technical teams to identify machine learning solutions for real-world problems. Collected, cleaned, and analyzed data using Python, SQL, Pandas, and NumPy to create datasets for model development. Built and fine-tuned machine learning models for customer segmentation, demand forecasting, fraud detection, and churn prediction. Performed feature engineering, data transformation, and model validation to improve accuracy and performance. Evaluated models such as Random Forest, XGBoost, Logistic Regression, and Decision Trees to select best performers. Developed FastAPI services to integrate models with web applications and other systems, and deployed/monitored models using Azure Machine Learning for production reliability. Collaborated with data engineering, software development, and QA teams across development, testing, deployment, and post-production support.

Delivered statistical analysis and predictive modeling for business reporting and analytical initiatives. Conducted exploratory data analysis to identify trends and actionable insights. Built regression and classification models using Python and R to support decision-making. Wrote SQL queries and prepared datasets for enterprise reporting. Performed data cleansing, preprocessing, and feature selection, working with XML, CSV, JSON, and relational datasets to support model development for classification and regression use cases.