There’s a new RAG technique every other week. Hybrid Search. Rewriting. Re-ranking. Agents. And while each solves something important, it’s easy to get overwhelmed — or worse, stuck trying to memorize patterns without knowing when to use them. So let’s slow down, break things

What we just experienced was true baseball. It’s a cruel, beautiful, joy filled, gut wrenching game. We saw incredible highs during the regular season and we saw incredible lows. We saw Julio look like a AAA hitter for 2 months and a true superstar for the second half. We

Welp. That one’s gonna hurt for a while. Love love love this team, though. Bring back Naylor and let’s run this shit back.

Look ma, I made it!

And we're live! Come join us:

If you're hacking around on AI projects and have questions about where the market is going and what trends we're seeing at @UnstructuredIO, check out the 🧵 below.

There's still time to sign up for today's webinar! 👇

The Document AI space has seen a fundamental shift in the past year. Everyone—from scrappy startups to established players—has pivoted from custom supervised models to wrapping the same handful of closed-source multimodal models. Yet, despite the fact we're all using essentially

Why are complex tables so hard to parse? OCR can detect characters, and some newer models can even handle simple tables. But once you introduce blank cells, multi-row headers, or nested structures, OCR quickly falls short. Rows and columns lose their positionality, context

At @UnstructuredIO, we often get the question "how well do you perform on scanned forms that include handwriting?" These types of documents are notoriously among the most difficult types of documents to ingest cleanly and reliably, yet they remain ubiquitous across many

Remember when extracting data from complex tables felt like digital archaeology? Messy. Painful. Incomplete. We do. That’s why we’ve devoted years of R&D to table transformation, turning one of document AI’s hardest challenges into a core strength. 1/🧵

In our latest webinar, we showed how Unstructured acts as Grand Central Station for GenAI pipelines—connecting sources, parsing content, chunking, and embedding to deliver AI-ready outputs. From VLM partitioning to contextual chunking and embeddings, we covered the exact way ETL

🏆 We’re ISO 27001 certified! We’re proud to announce that Unstructured is now officially ISO 27001 certified! This certification reflects our unwavering commitment to safeguarding customer data while delivering trusted solutions for transforming complex, unstructured data into

📣 Now Available: MLK Assassination Files — Structured and Searchable Following the declassification of over 6k documents and 240k pages related to the assassination of Dr. Martin Luther King Jr., Unstructured has released a machine-readable corpus built from these files. Many

What the fuuuuuuuuu

Unbelievable corruption here

It also shouldn’t be lost on anyone that Trump ordered these raids in large liberal cities. Places where his staff knew people would resist and where you could have “warzone” narratives that Fox News will eat up. There are no raids on the countless farms filled with

Lose $10T of market capitalization to fix a $1T trade deficit. 8D chess.

Wow. Unstructured just made Fast Company's list of the Top 50 Most Innovative Companies in the World—coming in at #24! 🔥 When we started, we believed that if AI was going to change the world, it needed better data. Fast forward, and we’re now helping some of the biggest

Not surprisingly, the Trump and GOP budget has nothing about no tax on tips, nothing on no tax on overtime, and nothing on no tax on social security. It does have a $4.5 trillion tax cut for billionaires however.