video1.0<iframe src="https://www.loom.com/embed/bfdb15672ff04dcf83472d41e9cbbfe8" frameborder="0" width="1728" height="1296" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>12961728Loomhttps://www.loom.com12961728https://cdn.loom.com/sessions/thumbnails/bfdb15672ff04dcf83472d41e9cbbfe8-4f82363d30421a18.gif222.336In this video, I walk you through how to effectively use pipelines in PromptHub, focusing on our product feedback classifier. I demonstrate how the pipeline integrates evaluations, datasets, and raw versioning to track performance during merge requests. Each time a merge request is made, the pipeline runs evaluations on a dataset of five test cases to ensure we maintain performance standards. I also highlight the importance of running guardrails to prevent issues like including secrets or profanity in prompts. Please make sure to implement these practices in your workflow to enhance our product's reliability.