aiden brotman

turning documents into data

extend:

[aiden brotman @brotdog]

@kushalbyatnal Best bet I made in a long time

[aiden brotman @brotdog]

nothing like a man sitting in a sofa on a subway car

[aiden brotman @brotdog]

@bolau_ @ExtendHQ @chaneratextend @DVarneyExtend these are AWESOME

[aiden brotman @brotdog]

the iphone duo chokehold

[aiden brotman @brotdog]

@ebadgio_ nice easter egg

[Kushal Byatnal @kushalbyatnal]

best part about moving into a new office is getting to throw a banger office warming party on the Brex card

Extend NYC 08.27.26

[aiden brotman @brotdog]

nyc is so LOUD

[aiden brotman @brotdog]

@BajorJoseph Very cool joe!!

[aiden brotman @brotdog]

Immediately made me think of:

[OpenAI @OpenAI]

This is GPT-6 Astra.

Anything you can do on a computer, Astra can do for you. Fast.

[aiden brotman @brotdog]

@kushalbyatnal even more affordable and accurate? what can't extend do?

[aiden brotman @brotdog]

lighter than a feather and it does parsing better!

[Kushal Byatnal @kushalbyatnal]

today, we're launching Light Parse

Extend can now parse documents up to 70% cheaper than other solutions

- light parse starts at $0.00625 per page, and scales down significantly with volume discounts

- it's SOTA on Databricks OfficeQA Pro (a 3rd party agent benchmark) among all non-agentic parsing providers

our ML/AI team trained smaller and more efficient layout, table, and form vision models to enable significant cost reduction without impacting accuracy on many types of documents

live today for all users!

[Kushal Byatnal @kushalbyatnal]

today, we're launching Light Parse

Extend can now parse documents up to 70% cheaper than other solutions

- light parse starts at $0.00625 per page, and scales down significantly with volume discounts

- it's SOTA on Databricks OfficeQA Pro (a 3rd party agent benchmark) among all non-agentic parsing providers

our ML/AI team trained smaller and more efficient layout, table, and form vision models to enable significant cost reduction without impacting accuracy on many types of documents

live today for all users!

[aiden brotman @brotdog]

when the parkday delivery arrives

[aiden brotman @brotdog]

@the_doc_doctor I just laughed so hard

[aiden brotman @brotdog]

@bolau_ 👀

[aiden brotman @brotdog]

@bolau_ so fire

[aiden brotman @brotdog]

Getting to walk through Union square market every morning is peak

[aiden brotman @brotdog]

@victordubeux oh my days

[aiden brotman @brotdog]

This is awesome -- really cool example of what happens when you can turn messy, scientific docs into something structured that researchers can actually build on.

[Victor Dubeux @victordubeux]

Decided to open source my weekend side project! S7 uses Extend API + LLMs to query genetic association data from GWAS/exome papers and reconcile it all into one clean, queryable schema with citations.

Inspiration came from watching Extend helping dozens of LiSci R&D teams handle nasty genetic tables that are pretty unqueryable at scale.

The stack is Pydantic + SQLite + FastAPI, in addition to Extend + your LLM of choice (I did Claude + GPT, dual-model).

[aiden brotman @brotdog]

@bolau_ @jordanalexmeyer @ExtendHQ love this