video1.0<iframe src="https://www.loom.com/embed/d20ec2f46ff5479e8d5fa9e1e2cc1765" frameborder="0" width="1920" height="1440" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>14401920Loomhttps://www.loom.com14401920https://cdn.loom.com/sessions/thumbnails/d20ec2f46ff5479e8d5fa9e1e2cc1765-31c18402a73866eb.gif216.875Hi, I'm Venu Madhav, and in this video, I introduced PipelineGuard, an AI agent designed to automatically triage failing data pipelines in seconds, saving engineers between 45 minutes to 2 hours per incident. I demonstrated how it utilizes Elasticsearch tools to analyze logs and search engineering runbooks, providing a complete triage report with root causes and fixed steps. I created two custom tools within the AgentBuilder to enhance its functionality, ensuring it operates efficiently. I encourage you to explore how PipelineGuard can transform messy data operations into reliable automation.