Madhu Samhitha Prayaga

Data Scientist with hands-on experience building end-to-end ML pipelines, regression and tree-based models, and feature engineering systems on real-world datasets. Proven ability to design scalable data pipelines, perform rigorous model validation, and derive actionable insights from structured and unstructured data. Currently expanding expertise in DevOps and production-level MLOps to bridge model development with deployment.

Eager to contribute to high-scale, low-latency real-time prediction systems where model performance directly impacts customer experience.

Data Science & Machine Learning Intern at Labmentix (2025-01 – 2025-08)

• Architected and deployed end-to-end ML pipelines in Python covering data ingestion, feature engineering, model training, and serving, ensuring system reliability and production-readiness.
• Performed advanced feature engineering on structured and unstructured datasets, extracting signals that improved model prediction accuracy.
• Applied regression and tree-based models (XGBoost, Random Forest) for forecasting tasks; conducted hyperparameter tuning and cross-validation to optimize performance.
• Conducted NLP and EDA on large datasets to extract actionable insights informing business forecasting strategies.
• Designed database schemas to store intermediate and final data products, enabling efficient querying and data integrity for downstream systems.
• Contributed to model monitoring setup to track performance post-deployment and flag data anomalies.

Data Analytics Intern at Trainity (2024-11 – 2025-01)

• Designed advanced SQL queries to extract and analyze customer data trends, building foundational datasets for reporting and predictive modeling.
• Developed and automated Tableau dashboards to visualize key business metrics, improving analytics efficiency for stakeholders.
• Automated Excel report generation using Python, reducing manual data processing time by over 80%.

Associate Operator – Geospatial Data Validation at Uber (2023-10 – 2024-03)

• Validated and cross-referenced large volumes of location-based data from multiple unstructured sources (social media, government records) to build a high-accuracy geospatial master dataset.
• Applied keyword-based research and qualitative data synthesis to verify geographic coordinates and resolve data discrepancies for operational mapping.
• Maintained strict data quality standards across high-volume datasets, ensuring reliability for downstream business intelligence and real-time mapping operations.
• Gained exposure to spatial data concepts including geo-coordinate verification and location signal accuracy — directly relevant to spatial feature engineering for ML models.

Programmer Trainee – Automation Testing at Cognizant (2022-10 – 2023-03)

• Designed and automated test scripts using Java and Selenium WebDriver to validate application functionality and data integrity across modules.
• Validated JSON/XML API responses, ensuring accuracy of backend data pipelines and system reliability.

Associate Software Engineer at Capgemini (2021-06 – 2022-03)

• Processed and resolved customer issues through complex workflows in ServiceNow; leveraged internal query tools to reduce average ticket handling time.
• Identified and documented SOP improvements to streamline ticket resolution processes.

Bachelor of Science in Statistics – Sri Venkateswara University, Tirupati (2017 – 2020)