Software · AI & automation [SO · DOCUMENT AI]
Reading documents
PDFs, scans and emails turned into clean data.
Pipelines that read invoices, contracts, forms and emails and turn them into structured records you can search, check and report on.
Reading documents
- Built with
- Python, OCR, Claude, Postgres
- You own
- Source code, the repository and the accounts it runs on. No per-seat licence to us.
- Where it runs
- Your own cloud account, or ours if you would rather not run it.
- Scoping
- One call, then a written scope. We say no when it is not ours.
Contact salesGet a quote
01 The terms
We quote the work after one call, in writing, against a scope you can hold us to. No published band, because a number without a scope is a guess — and no retainer that quietly renews.
What you get at the end is the thing itself and everything needed to keep it running without us. That is the deal on every job, whatever it is.
02 What we build
01 / 061
Extraction from PDFs, scans and email
02 / 062
Your own rules
Validation against your own rules
03 / 063
A review queue for anything uncertain
04 / 064
Search everything
Search across every document
05 / 065
Export anywhere
Export to your database or spreadsheets
06 / 066
Source-to-record trail
Audit trail from source to record
03 Examples
Three things we could build for you here. They show the range; they are not a client list.
Supplier invoices captured into accounts
Contract terms extracted into a register
Shipping and trade documents turned into a ledger
Scroll sideways for the rest3 examples
04 Is this you?
Who asks for this
- Teams re-typing documents
- Finance and operations back offices
What we build it with
05 Next to this
Next Reading documents
Tell us what it has to do.
No form, no sales sequence. Write two lines about the problem and you will get a person who has read them.