Sai Raja Ramya Bhavana Thota

About

Data/ML practitioner with experience applying NLP and classical ML to large-scale customer data (2B+ records) for segmentation, prediction, and survey-text classification, delivering measurable business impact (~18% engagement efficiency). Has hands-on entity resolution across multi-source datasets and has built embedding-based semantic search using SentenceBERT + a vector database with domain fine-tuning (~20% relevance improvement), plus production workflow experience with Spark/Airflow and cloud tooling (AWS/Azure).

Key Strengths

• Scaled ML/NLP data processing on 2B+ structured and unstructured records
• End-to-end customer segmentation + propensity modeling (K-means, logistic regression, random forest)
• Rigorous model selection/validation (elbow + silhouette, k-fold CV, ROC/AUC, precision/recall, F1/F4)
• Business-impact delivery: dashboards in Tableau/R Shiny and ~18% engagement efficiency improvement
• Entity resolution across multi-source customer data using hybrid matching + TF-IDF similarity with manual QA
• Built semantic search/data linking using SentenceBERT embeddings + vector database; ~20% relevance lift
• Production-grade workflow engineering (modular pipelines, logging, config, unit tests; Spark/cloud scaling)
• Pragmatic prototype-to-production decision-making with deployment via Docker/Azure ML

Experience

Education

Skills

Languages