Canary releases, a modern software delivery technique, facilitate stress-free rollouts of new application versions by deploying a new release alongside an existing version and directing a small fraction of production traffic to it. This method allows for incremental updates if the new version proves stable. Utilizing the Codefresh-Kubernetes integration, this process involves building application and test images, deploying the canary using Kubernetes deployment tools, and conducting end-to-end integration tests to verify the canary's health. If successful, the canary is exposed to 10% of production traffic, and upon passing quality gates, a Codefresh pipeline is triggered to update all production pods. The example provided focuses on deploying an ornithology data management service using Python and a MongoDB backend, employing trunk-based development and delivery patterns with two Codefresh pipelines. The deployment strategy emphasizes automated verification processes, traffic distribution, and a rolling update mechanism for production pods, highlighting the importance of defining quality gates and integrating observability platforms for effective monitoring and decision-making during the release process.