Developed a robust Docker execution environment for the pantsbuild/pants repository, focusing on improving runtime resilience and reducing operational disruptions. Leveraging Python and Rust, the work introduced structured error handling for container-related issues and implemented an automatic retry mechanism to address transient container unavailability or removal. This approach minimized the need for pantsd restarts by enabling the system to recover gracefully from common Docker failures. Expanded test coverage validated recovery paths across various container states, ensuring reliability and maintainability. The project drew on skills in concurrency, system programming, and testing to deliver a more stable and fault-tolerant execution environment.