Most enterprise data is "dark data"—unstructured PDFs, scans, and media files scattered across servers that usually require fragile ETL pipelines before AI can use them.
Google Cloud Developer Relations Engineers Smitha Kolan and Luke Schlangen explain how Smart Storage and Knowledge Catalog make unstructured data ready for AI agents:
* *Zero fragile ETL pipelines:* Skip complex extract, transform, and load pipelines just to index scattered files.
* *In place classification:* Smart Storage automatically identifies and classifies unstructured PDFs, scans, and media right where they live.
* *Mapped in Knowledge Catalog:* Turns unindexed storage into structured context that Gemini agents can immediately reason over.
#Shorts #GoogleCloud #GeminiAtWork #SmartStorage #DataCloud #AIAgents
Speakers: Smitha Kolan, Luke Schlangen
Products Mentioned: Gemini, Smart Storage, Knowledge Catalog
