{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/f798e2408f9e4ed5afcf453b16140944\" frameborder=\"0\" width=\"1148\" height=\"861\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":861,"width":1148,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":861,"thumbnail_width":1148,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/f798e2408f9e4ed5afcf453b16140944-ec5c3182751e08d4.gif","duration":914.328,"title":"Databricks-IAB-AAMP-Buy-Sell Demo","description":"This Loom explains the IAB Agentic Advertising AMP protocol on Databricks and how it enables end to end agentic media buying on open standards. The speaker describes a current problem where buying media is manual, siloed, and relies on bespoke integrations, which agents cannot bridge without shared language and trust. AMP standardizes how buyer and seller agents transact using open language, open direct for transactions, open RTB for deals, and a registry for identity and trust. Built on Databricks, the solution uses LakeBase for agent state, Model Serving on Cloud, Unity Catalog for data governance, MLflow for decision tracing, and Databricks apps to host buyer, seller, and registry identities, with a self deployed accelerator to follow."}