Goodbye scraping hacks. Goodbye broken PDF tools. You can now...
EyeingAI@EyeingAI
26 views
Sep 24, 2025
~2 min read
Advertisement
1
Goodbye scraping hacks.
Goodbye broken PDF tools.
You can now extract anything from PDFs... text, tables, charts, even multi-column layouts & get clean Markdown or JSON, instantly.
Examples + how I’m using it: 🧵
Goodbye broken PDF tools.
You can now extract anything from PDFs... text, tables, charts, even multi-column layouts & get clean Markdown or JSON, instantly.
Examples + how I’m using it: 🧵
2
Ok, so I used to spend way too much time pulling data out of PDFs.
Now I just tell this AI what I need and it builds the extractor for me.
No code.
No setup.
Just a normal conversation.
Doctly’s new Extractor Studio is kinda nuts.
doctly.ai
Now I just tell this AI what I need and it builds the extractor for me.
No code.
No setup.
Just a normal conversation.
Doctly’s new Extractor Studio is kinda nuts.
doctly.ai
0:19
3
Here’s how it works:
– You upload a PDF... could be 100+ pages, complex as hell.
– Doctly then runs a page-wise model selection. It doesn't use one generic parser.
– It actually detects what kind of layout each page has, and uses the right engine accordingly.
Smart af.
– You upload a PDF... could be 100+ pages, complex as hell.
– Doctly then runs a page-wise model selection. It doesn't use one generic parser.
– It actually detects what kind of layout each page has, and uses the right engine accordingly.
Smart af.
0:37
4
And it’s not just a one-click tool either.
It’s a full-on dev environment.
You can test with multiple PDFs, tweak your prompts, track versions, and even compare JSON outputs side-by-side before pushing anything live.
It’s a full-on dev environment.
You can test with multiple PDFs, tweak your prompts, track versions, and even compare JSON outputs side-by-side before pushing anything live.
0:31
5
If you’re dealing with critical data, there’s an Ultra mode.
It runs a multi-layer LLM setup to make sure the output’s accurate, stable, and consistent, every time.
No more broken JSON. No surprises. Just clean results you can trust. 🔥
It runs a multi-layer LLM setup to make sure the output’s accurate, stable, and consistent, every time.
No more broken JSON. No surprises. Just clean results you can trust. 🔥
0:50
6
And yes... it has an API.
So now I have a tiny script that:
– grabs incoming PDFs
– feeds them to Doctly
– auto-dumps clean outputs into my workspace
So now I have a tiny script that:
– grabs incoming PDFs
– feeds them to Doctly
– auto-dumps clean outputs into my workspace
7
My favorite part?
You can export your results into:
– Markdown (for Notion, blogs, docs)
– JSON (for devs, automations, databases)
– Raw structured text for copy-paste
Which means:
No more copy-clean-format loops.
You can export your results into:
– Markdown (for Notion, blogs, docs)
– JSON (for devs, automations, databases)
– Raw structured text for copy-paste
Which means:
No more copy-clean-format loops.
8
This is a new superpower for anyone who builds with data.
Check out the AI Extractor Studio and see for yourself.
doctly.ai
Check out the AI Extractor Studio and see for yourself.
doctly.ai
9
Done!
If this thread helped you,
👉 Repost it to help someone else.
👉 Follow @EyeingAI – I break down AI every day.
If this thread helped you,
👉 Repost it to help someone else.
👉 Follow @EyeingAI – I break down AI every day.
EyeingAI@EyeingAI
Goodbye scraping hacks.
Goodbye broken PDF tools.
You can now extract anything from PDFs... text, tables, charts, even multi-column layouts & get clean Markdown or JSON, instantly.
Examples + how I’m using it: 🧵
Goodbye broken PDF tools.
You can now extract anything from PDFs... text, tables, charts, even multi-column layouts & get clean Markdown or JSON, instantly.
Examples + how I’m using it: 🧵
