Tell us the problem you want to solve
Thank you. We have your message and will reply within one business day.
Automate the routine work so your teams can hit their goals
We automate, streamline, and fine-tune work such as fraud detection, finance operations, and engineering design. When a finance team can keep up with every customer deduction, the company stops losing revenue to over-deductions. Whatever the industry, we build each system around your data, your people, and your business requirements.
What we build
Finance and back office
Matching supplier statements against your ledger, tracking deductions and short payments, and flagging the exceptions your finance team needs to act on. Agents prepare the work (a reconciliation, a case file, a draft report) and hand the decision to a person, with language-model usage and its cost visible and capped.
Fraud and AML
Models that catch fraud, money laundering, and suspicious transactions that rules miss, and keeps catching them as behaviour changes.
Credit decisions
Credit scoring and approval models on transaction and credit bureau data, with the explanations your credit committee and regulator need.
Documents and knowledge
Extraction from statements, claims, contracts, and reports into a checked data layer that records the source of each fact, so answers can be traced and audited.
Engineering design
Systems that read drawings and diagrams, draft bills of materials and follow-on documents, and carry a component change through every document that depends on it, for your engineers to review.
Demand forecasting
Daily or weekly forecasts of demand per product and site, from sales history, stock, promotions, public holidays, paydays, weather, and other signals. Each forecast sets how much to make, order, and hold, so that you meet demand with less waste and less cash tied up in stock. We measure all three against your current baseline.
AI strategy
Senior practitioners who help your leadership choose use cases, set realistic cost and accuracy targets, and build the internal capability to run them. Where data must stay in South Africa, we fine-tune small models that match large general-purpose models on your task and run them on your own infrastructure.
Built for regulated and critical work
Explainable by design
Each score comes with the reasons behind it. Where a decision must follow fixed, auditable rules, we derive those rules from the model.
Your data stays yours
Systems run in your cloud or on your hardware. You own the code, the models, and the documentation. You pay no per-seat or per-query licence fees.
People stay in control
Investigators, analysts, credit officers, and engineers decide; the system ranks cases, explains its scores, and drafts case files and reports for them to check.
Local and senior
We are a South African team in your time zone, led by people who have built production machine learning since 2014.
AI that costs less to run as it is used more
Frontier models are cheap to start with, but their cost rises with every user and every question. We engineer the data, pick the most efficient and effective model for the accuracy target, and run it on infrastructure you control, so the running cost stays flat as use grows.
| General-purpose | DataProphet-engineered | |
|---|---|---|
| Set-up cost | Low | Higher |
| Cost as users grow | Rises with every user | Flat until a capacity upgrade |
| Cost as questions get harder | Rises steeply | Flat until a capacity upgrade |
| Changing model | Often means starting over | Swap the model, keep the data layer |
Case studySequence learning for fraud detection on a crypto exchange
OVEX's rules caught three quarters of fraud and raised alarms that were mostly false. We replaced them with models that read the order and timing of each account's activity.
We trained two models on OVEX's history, one on how fraudsters behave and one on how honest customers behave, and the system flags an account when it looks more like the first. Now in production, the system includes a dashboard that shows investigators when an account's behaviour changed.
Read the case study →“We partnered with DataProphet for their deep machine-learning expertise and they've built a system tailored to our platform, giving us a far more sophisticated way to identify fraud as it evolves.”
Selected work
How an engagement starts
Readiness assessment
We review your data and goal and deliver a readiness report, a workshop with your team, and a delivery plan.
Build and validate
We build a working model and measure it against your own KPIs. You then decide whether to go on.
Production and improvement
We deploy, integrate, and keep the models current, and train your team to run them.