Skip to main content

SOLUTIONS · FINANCE

Financial ML, solved end-to-end

Real forecasting and risk problems worked start to finish on relational data — accounts, counterparties, entities and the links between them. Each solution ships a live model and a shareable dashboard, with your data isolated to your account.

Talk to our finance team Privacy & security

Case studies

Solutions your teams can ship

Each one is a working example — the problem, the trained model, the questions you can ask it in plain English, and the dashboard. Filter by product track or ML task.

Track

ML task

Data Science Regression

Predict a film's box office from its production data

Estimate worldwide revenue for a film that does not exist yet — from budget, runtime, language, director and the cast/company/genre graph around it. A worked example of a relational model on 9 linked tables, and an honest account of how much of the signal is just budget.

A revenue estimate for a film that has no row yet — and a ranking you can green-light against.

Sep 3, 2026 · 10 min read Read the solution →

On the roadmap

More finance solutions in progress

Data Science Binary classification
Coming soon

Fraud detection on the transaction graph

Score transactions against the graph of accounts, devices, merchants and counterparties — the multi-hop structure a flat feature table throws away.

Data Science Binary classification
Coming soon

Credit scoring from your own history

Complement bureau scores with signals from the repayment and transaction history you already hold.

Data Science Multi-class classification
Coming soon

Transaction categorisation

Label raw transaction descriptors with canonical categories for PFM, accounting and tax products.

Data Science Anomaly detection
Coming soon

Anomalous transfer patterns

Surface structured-transfer and layering behaviour without needing a labelled fraud history to start from.

Want one of these worked on your data first? Tell us what you're solving →

Point it at your own ledger

Connect your tables, pick the column you want predicted, and train — no pre-joining, no feature engineering, no infrastructure to run.

Start free Talk to our team