About
A practice, not a product company
What we do
One narrow problem, done thoroughly
We keep base models — models already trained on floor plans — for the things every drawing has: walls, rooms, doors, windows, tags, hatch patterns. They are the starting point for every engagement and the thing we run in the free analysis, which is why a project begins weeks in rather than at an empty model and a year of labelling.
What a client buys is a model tuned on their own drawings, measured against their own manual results, running where their team or their product needs it.
How fine-tuning worksWho we are
Part of AxcelerateAI
- Offices in Austin, TX and Lahore, PK
- NDA signed with every client before drawings are shared
- Client drawings are never used to train models for anyone else
- Clients own the weights trained on their data
Contact goes to consult@axcelerate.ai, which reaches an engineer rather than a queue.
How we work
Things we would rather say up front
- We do not quote an accuracy figure before seeing your drawings.
- If a job is not worth automating, we will say so rather than scope it.
- Information that is not on the drawing cannot be extracted from the drawing.
- Nothing we build signs anything off — a person reviews every finding.
- Where a project stopped at a proof of concept, our write-up says so.
You can see how that reads in practice on the work pages.
Get in touch
Tell us what is on your drawings
Describe the sheets you get and what you need off them. We will tell you what our models already handle, what needs fine-tuning on your drawings, and what is not worth automating at all.
- Reply time
- Within 1 business day
Would rather show than tell? Send one sheet for a free analysis