Worked on upgrading the Dspy library to version 2.6.0 within the Arize-ai/openinference repository, focusing on aligning both configuration and main instrumentation code with the latest API. The approach involved careful code refactoring in Python, removing unused imports, and updating module references to ensure seamless compatibility with the new library version. Emphasized dependency management to validate repository-wide compatibility, reducing the risk of regressions and preparing the codebase for future enhancements. This work improved maintainability and stability, ensuring that the project can leverage new features and fixes introduced in the updated Dspy library while maintaining clean and organized code.