Shubhangi Raj developed automated model upload validation for the Meesho/BharatMLStack repository, focusing on enhancing deployment reliability and CI quality gates. She implemented backend validation logic in Python, ensuring that uploaded models meet configuration requirements, including warmup settings in config.pbtxt files. Her approach included adding comprehensive unit tests, resolving static analysis and lint issues, and improving code maintainability by removing redundancies and unnecessary comments. Shubhangi also validated logger and print statements within backend models to improve observability and reduce runtime errors. Her work demonstrated depth in backend development, model validation, and GCS integration, primarily using Go and Python throughout the project.