Matt Robinson

software engineer @ unstructured-io (

[VentureBeat @VentureBeat]

Announcing our #VBTransform Innovation Showcase finalists: @ArmillaAI ,@yellowdotai , AnswerAI, @SupportLogicInc, @clearmlapp @SkyflowAPI @UnstructuredIO @arizeai. Two more finalists will be selected by attendees at VBTranform in SF July11&12! Read more: https://t.co/lPdJmWpI00

[Unstructured @UnstructuredIO]

Awesome to see Unstructured make the front page of Hacker News. Please upvote "Demystifying Text Data with the Unstructured Python Library"!!

https://t.co/dGOYh44VUw

[Unstructured @UnstructuredIO]

Great piece by @BornsteinMatt and @rajko_rad on the emerging architectures for LLM applications. At Unstructured we're building the first ETL platform for LLMs. We want to make it easy for data scientists to connect, transform, and stage their data for vector strorage/fine tuning

[Matt Robinson @mrobinson0623]

@Klaudioz @langchain @UnstructuredIO Created an issue to track: https://t.co/yTh1fGC9Si

[Matt Robinson @mrobinson0623]

@Klaudioz @langchain @UnstructuredIO In the meantime though you could run the downloading code from GoogleDriveLoader yourself and pass the downloaded file to UnstructuredExcelLoader

[Matt Robinson @mrobinson0623]

@Klaudioz @langchain @UnstructuredIO Right now the download and load step occur in the same method in GoogleDriveLoader, so I don't think there's an easy way to combine the two loaders at the moment. Will create an issue to address that though

[Crag Wolfe @ai_4_all_docs]

https://t.co/vecdzM6no6 is great for processing individual raw documents. Now it's easy to process many at once and store the structured data for all of your downstream use cases.

Just point to an s3:// bucket and let the tool churn away:
https://t.co/woB7yFgz45