We couldn't find an API that understood construction drawings, so we built one

getanchorgrid.com

1 point·wcisco17··4 comments

My co-founder and I got pulled out of e-commerce into construction. What we found was a document format that breaks every tool you throw at it — OCR, Textract, Mistral, SVG conversion. Schedules with no consistent structure, elevations no CV model understood. Wrote about the two walls we hit and why we think construction needs infrastructure not another UI.

http://getanchorgrid.com/developer

There is some video of the screen while using your product in https://www.getanchorgrid.com/developer Do you have a more detailed example?

What specifically would be most useful — a door detection example, spec parsing, or elevation detection?

You can use it like this --

Upload a pdf first: curl -X POST https://api.anchorgrid.ai/v1/documents \\ 90 + -H "X-API-Key: <your-api-key>" \\ 91 + -F "file=@floor-plan.pdf"

then run the door detection for the pdf: curl -X POST https://api.anchorgrid.ai/v1/drawings/detection/doors \ -H "X-API-Key: <your-api-key>" \ -H "Content-Type: application/json" \ -d '{ "document_id": "550e8400-e29b-41d4-a716-446655440000", "page_numbers": [1, 2, 3] }

I also added an example PDF (just so you can interact with the results live)

https://www.getanchorgrid.com/developer/docs/endpoints/drawi...

Reply on news.ycombinator.com