ⓘ Read-only hosted preview!! no analysis or DataHub mutation is executed. Clone the repository to run the live workflow.

Provenance Logo
Production ML agent View repository Run locally with DataHub →
← All reports

Recorded BLOCK evidence

Device trust score contract change

device-trust-range-change--block--20260807T140453Z-857303.md

Provenance Impact Report

  • Run ID: 20260807T140453Z-857303
  • State: PUBLISHED
  • Started: 2026-08-07 14:04:53.120572+00:00
  • Completed: 2026-08-07 14:04:58.493510+00:00

Proposed Change

  • Asset: urn:li:dataset:(urn:li:dataPlatform:kafka,payments.authorized_transaction_events,PROD)
  • Source column: device_trust_score
  • Scenario: device-trust-range-change
  • Request: The device-risk producer will change device_trust_score from a 0–100 range to a 0–1 range without updating downstream expectations.
  • Trigger event: evt-dd8cd270-a249-4a94-9958-1328a6363c9d
  • Trigger source: device-risk-deployment-agent

DataHub Context (through MCP)

  • Source: urn:li:dataset:(urn:li:dataPlatform:kafka,payments.authorized_transaction_events,PROD) (payments.authorized_transaction_events)
  • Selected field: device_trust_score
  • Owners: urn:li:corpGroup:payment-fraud-monitoring
  • Tags: Production, PaymentFraud, RealTimeML

Complete lineage

Hop Upstream Downstream Relationship
1 urn:li:dataset:(urn:li:dataPlatform:kafka,payments.authorized_transaction_events,PROD) urn:li:dataset:(urn:li:dataPlatform:postgres,fraud.realtime_payment_features,PROD) lineage
2 urn:li:dataset:(urn:li:dataPlatform:postgres,fraud.realtime_payment_features,PROD) urn:li:mlFeature:(realtime_payment_risk_features,device_trust_score) lineage
3 urn:li:mlFeature:(realtime_payment_risk_features,device_trust_score) urn:li:mlModel:(urn:li:dataPlatform:mlflow,payment_fraud_detector_v3,PROD) lineage

Affected models

  • urn:li:mlModel:(urn:li:dataPlatform:mlflow,payment_fraud_detector_v3,PROD) ➔ payment_fraud_detector_v3

Evidence URNs

  • urn:li:dataset:(urn:li:dataPlatform:kafka,payments.authorized_transaction_events,PROD)
  • urn:li:dataset:(urn:li:dataPlatform:postgres,fraud.realtime_payment_features,PROD)
  • urn:li:mlFeature:(realtime_payment_risk_features,device_trust_score)
  • urn:li:mlModel:(urn:li:dataPlatform:mlflow,payment_fraud_detector_v3,PROD)

Experiment Plan

  • Target asset: urn:li:dataset:(urn:li:dataPlatform:kafka,payments.authorized_transaction_events,PROD)
  • Rationale Explanation: Reproduce the proposed device-score range change across the full evaluation population and measure its effect on the downstream fraud model.
  • Expected signals: AUC drop, prediction flip rate, mean absolute probability shift
# Operator Column Fraction Value
1 scale_numeric device_trust_score 1.0 0.01

Observed Results

Experiment 1: scale_numeric

  • Column: device_trust_score
  • Duration (how long the mutation and mutated inference took)[ms]: 17.902000000049156
  • Error: n/a
Metric Observed value
baseline_auc 0.8092934782608696
mutated_auc 0.7495108695652174
auc_drop 0.05978260869565222
prediction_flip_rate 0.974
mean_absolute_probability_shift 0.8007552497510132
inference_error_rate 0.0
  • Rows selected: 500 / 500
  • Before summary: {"maximum": 99.42, "mean": 72.66107999999998, "minimum": 22.51, "missing": 0}
  • After summary: {"maximum": 0.9942000000000001, "mean": 0.7266107999999999, "minimum": 0.22510000000000002, "missing": 0}

Deterministic Decision

  • Status: BLOCK
  • Summary: Blocked by 2 deterministic safety rule(s).

Triggered policy rules

Experiment Metric Observed Rule Result Reason
1 auc_drop 0.05978260869565222 > 0.05 BLOCK AUC drop 0.059783 exceeded block threshold 0.050000
1 prediction_flip_rate 0.974 > 0.15 BLOCK Prediction flip rate 0.974000 exceeded block threshold 0.150000

Publication

  • Local JSON: mockup_data/artifacts/runs/20260807T140453Z-857303.json
  • Local Markdown: mockup_data/artifacts/runs/20260807T140453Z-857303.md
  • Remote succeeded: yes
  • DataHub tag: Provenance-BLOCK
  • DataHub document: urn:li:document:shared-1f1429aa-2bb7-42c8-b76b-d717065e3b58
  • Remote error: n/a