LatePay — know which invoices will ghost you, before they do 💸

Description and use

LatePay helps freelancers predict late payments on invoices, offering specific actions for each risk level.

Problem

Freelancers struggle with delayed payments, risking cash flow when a single late payer stalls revenue for a month.

Outcome

LatePay provides a risk‑based prioritization of invoices, reducing the likelihood of cash‑flow disruptions by guiding freelancers on when to chase, request upfront payment, or send polite nudges.

Workflow

• 1. Enter invoice details (amount, client tenure, prior late rate, etc.) into the form.
• 2. TabPFNClassifier performs in‑context inference on a 1,000‑row simulated dataset to predict P(paid >30 days).
• 3. The system outputs a risk band and a clear action (e.g., chase now, request upfront, send nudge, or normal terms).
• 4. For batch scoring, upload a CSV of invoices; the tool returns a scored risk report with recommendations per row.
• 5. Users can review the static demo with precomputed predictions or import real invoice history for personalized context.

Evidence

LatePay — a late-payment risk predictor that scores every outstanding invoice with TabPFN, the open-source tabular foundation model, and tells you what to do about it: chase now, ask for partial upfront, send a reminder, or relax.

Source

TabPFN returns P(paid > 30 days past terms) as a risk band: HIGH / MEDIUM / ELEVATED / LOW.

Source

The model learned real, sensible patterns — a clean client scores 5.0% late risk, a serial late-payer 63.4% — but it has never seen a real invoice.

Source