THE QUESTION IT ANSWERS
Which samples are damaging which component?
- DETECT
01
The problem
Bad labels in training data cause silent model degradation. Retraining a model merely to find and fix those errors costs between $100,000 and $500,000. The Proofreader finds them before you retrain.
02
What it delivers
Attaches diagnostic readouts directly to a frozen model checkpoint. It returns a precise map of which labels in your dataset are objectively wrong and which are merely hard for the network to learn. The underlying model does not change; only the diagnostic insight is extracted.
03
Who it is for
A fit for
- Data leads
- Research informatics directors
- ML teams fine-tuning models on scraped, aggregated, or third-party data
04
How it works
The Proofreader is built from one capability of the ORMAS architecture. Each is a property of how the network trains, not a tool bolted on afterwards.
- 01DETECTIdentify which samples are damaging which component, and separate mislabelled from hard-but-correct.
05
In your industry
This is why ORMAS is one product rather than seven. The Proofreader is the same mechanism in every row below — only the customer's own vocabulary changes.
- Medical
- mislabelled scans
- Finance
- mislabelled events
- Biotech
- mis-annotated assays
- AI lab
- web-scrape noise
- SaaS
- mislabelled tickets
06
Markets using it
- AI TrainingSegments by scale
- Regulated FinanceSegments by industry
- Medical AISegments by industry
- Data ObligationSegments by obligation
- Defense & Safety-CriticalSegments by industry
07
Questions
No. ORMAS is one product under one licence, and The Proofreader is one of the seven things it is proven at. You turn on what your situation needs; there is no separate purchase, no separate integration and no separate contract.
The mechanism does not know what industry it is in. What changes between them is the noun, not the method — the table above is the same mechanism with each sector's own vocabulary substituted, which is why one product serves all of them.
On your infrastructure. The licence is on-premise and per-institution: nothing is sent to a third party, there is no hosted tier, and no usage metering phones home. In a hospital enclave or a controlled environment, that is a requirement rather than a preference.
Pre-book ORMAS with The Proofreader in scope.
Every engagement starts with one specific problem — a run that died, an audit that asked a question you could not answer, a dataset you cannot vouch for.
We reply within a few days. Every message is read by a person.
Pre-book a deployment
One licence, no price list. Terms are agreed per institution and written into the contract, and pre-booking fixes them ahead of general availability.
Useful to bring
- What you train, and roughly at what scale
- The obligation driving it — SR 26-2, the EU AI Act, Article 17, a PCCP, an internal policy
- What a failed run currently costs you to diagnose