video1.0<iframe src="https://www.loom.com/embed/065733f372fc4c2abcea2ae830de42d4" frameborder="0" width="1670" height="1252" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>12521670Loomhttps://www.loom.com12521670https://cdn.loom.com/sessions/thumbnails/065733f372fc4c2abcea2ae830de42d4-8021dbb8a9270b5d.gif420.203This Loom explains the improved CSV import and export experience in the Kernel platform using templates and AI agents. It shows how to import exported product data from catalog files by analyzing CSV contents, generating a draft mapping for fields like titles and context data such as dimensions and properties, previewing results, and running the import so products immediately classify into Kernel’s taxonomy for enrichment. It also demonstrates reusing the same imported structure for another file, such as a July collection, to import safely and quickly without redoing mappings. Finally, it covers importing from vendor unstructured data by running an AI agent to map supplier fields into existing Kernel attributes, including renaming a field from main category to product category, and reviewing side by side imports from core and supplier sources.