CHANDINI SAISRI UPPUGANTI
I'm a
I'm a graduate student in Software Engineering (Data Science) at San Jose State University with 4+ years of experience as a Data Engineer. I specialize in building scalable data pipelines, deploying ML/LLM-based solutions, and working across cloud ecosystems like AWS and Azure. My work spans legal AI tools, product ranking engines, and Retrieval-Augmented Generation systems—focused on delivering impactful, intelligent solutions.
Master of Science in Computer Software Engineering in Data Science
GPA - 3.8/4
Bachelor of Technology in Computer Science and Engineering
GPA - 3.9/4
Developed a pairwise classification model using fine-tuned MiniLM and BERT via HuggingFace Transformers and Unsloth to link argument-counterargument sections in legal briefs. Built preprocessing pipelines for JSON datasets and deployed an interactive Streamlit UI to visualize and explore predictions for legal research assistance.
Designed a hybrid product scoring system using NLP embeddings (BERT, Word2Vec) and graph-based similarity to personalize search relevance. Built and evaluated ranking models (Logistic Regression, SVM, Gradient Boosting), and optimized preprocessing pipelines to enhance model training by 30% and user engagement by 25%.
Pioneered development of a sophisticated Retrieval-Augmented Generation (RAG) model, integrating LLM with external databases, boosting poem retrieval precision by 45% and generating highly meaningful and contextually rich poetic content.
Developed a Sales Forecasting and Optimization System using ARIMA, SARIMA, and LSTM, improving forecast accuracy by 20% through EDA, trend analysis, and feature engineering (lag variables, moving averages, economic indicators). Optimized models with cross-validation and hyperparameter tuning, reducing RMSE and MAE for precise sales predictions.
Developed a multi-agent system using Model Context Protocol (MCP) and Agent-to-Agent (A2A) communication patterns. The system leverages AWS services to create a scalable architecture for distributed AI agents, enabling efficient context sharing and collaborative decision-making through innovative communication protocols.
