Detect Layered Laundering Patterns
“Which accounts will move more than $100K in total or execute more than 30 cross-border wires next month?”
Book a demo and get a free trial of the full platform: data science agent, fine-tune capabilities, and forward-deployed engineer support.
By submitting, you accept the Terms and Privacy Policy.
Loved by data scientists, ML engineers & CXOs at
A real-world example
Which accounts will move more than $100K in total or execute more than 30 cross-border wires next month?
Layered laundering uses two tactics: large aggregate volume (placement) and high-frequency small transfers (layering). AML systems look at each signal separately, missing accounts that split activity across both patterns. A compound prediction captures both tactics in one query.
How KumoRFM solves this
Graph-powered fraud intelligence
Kumo’s compound conditions combine multiple AML signals into a single prediction with OR logic. It captures both the placement pattern ($100K+ aggregate) and the layering pattern (30+ small transfers), catching accounts that split activity to stay below individual thresholds.
From data to predictions
See the full pipeline in action
Connect your tables, write a PQL query, and get predictions with built-in explainability — all in minutes, not months.
Your data
The relational tables Kumo learns from
Accounts
| account_id | account_holder | risk_score | country |
|---|---|---|---|
| A001 | Apex Corp | 72 | US |
| A002 | Offshore Ltd | 91 | Cayman Is. |
| A003 | Vega LLC | 35 | US |
Wire Transfers
| wire_id | account_id | amount | beneficiary_country | timestamp |
|---|---|---|---|---|
| W001 | A001 | 12,000 | UK | 2025-01-05 |
| W002 | A002 | 9,800 | Panama | 2025-01-12 |
| W003 | A002 | 9,700 | BVI | 2025-01-08 |
Write your PQL query
Describe what to predict in 2-3 lines — Kumo handles the rest
PREDICT SUM(WIRE_TRANSFERS.AMOUNT, 0, 30, days) > 100000 OR COUNT(WIRE_TRANSFERS.*, 0, 30, days) > 30 FOR EACH ACCOUNTS.ACCOUNT_ID
Prediction output
Every entity gets a score, updated continuously
| ACCOUNT_ID | TIMESTAMP | TARGET_PRED | True_PROB |
|---|---|---|---|
| A001 | 2025-02-01 | False | 0.18 |
| A002 | 2025-02-01 | True | 0.92 |
| A003 | 2025-02-01 | False | 0.07 |
Understand why
Every prediction includes feature attributions — no black boxes
Account A002 (Offshore Ltd)
Predicted: 92% laundering probability
Top contributing features
Wire transfer volume (30d sum)
$19,500
35% attribution
Cross-border wire count (30d)
28 wires
27% attribution
Account risk score
91
18% attribution
Beneficiary country diversity
Panama, BVI
13% attribution
Account country
Cayman Is.
7% attribution
Feature attributions are computed automatically for every prediction. No separate tooling required. Learn more about Kumo explainability
PQL Documentation
Learn the Predictive Query Language — SQL-like syntax for defining any prediction task in 2-3 lines.
Python SDK
Integrate Kumo predictions into your pipelines. Train, evaluate, and deploy models programmatically.
Explainability Docs
Understand feature attributions, model evaluation metrics, and how to build trust with stakeholders.
Bottom line: The OR condition captures two laundering patterns in one query. Cut false positives 60–70%, save 8,000+ investigator hours annually. Catch layering that single-signal rules miss.
Related scenarios
Explore more fraud predictions
Topics covered
One Platform. One Model. Predict Instantly.
KumoRFM
Relational Foundation Model
Turn structured relational data into predictions in seconds. KumoRFM delivers zero-shot predictions that rival months of traditional data science. No training, feature engineering, or infrastructure required. Just connect your data and start predicting.
For critical use cases, fine-tune KumoRFM on your data using the Kumo platform and Data Science Agent for 30%+ higher accuracy than traditional models.
Book a demo and get a free trial of the full platform: data science agent, fine-tune capabilities, and forward-deployed engineer support.
