ⓘ 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 PASS evidence

New digital-wallet channel rollout

new-digital-wallet-channel--pass--20260807T171510Z-10f593.md

Provenance Impact Report

  • Run ID: 20260807T171510Z-10f593
  • State: PUBLISHED
  • Started: 2026-08-07 17:15:10.508382+00:00
  • Completed: 2026-08-07 17:15:39.658262+00:00

Proposed Change

  • Asset: urn:li:dataset:(urn:li:dataPlatform:kafka,payments.authorized_transaction_events,PROD)
  • Source column: payment_channel
  • Scenario: new-digital-wallet-channel
  • Request: Five percent of payments will begin using a new payment_channel category named digital_wallet.
  • Trigger event: evt-6f2d23ab-4257-4cb7-89e8-5f35dba0e037
  • Trigger source: payment-platform-release-agent

DataHub Context (through MCP)

  • Source: urn:li:dataset:(urn:li:dataPlatform:kafka,payments.authorized_transaction_events,PROD) (payments.authorized_transaction_events)
  • Selected field: payment_channel
  • 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,payment_channel) lineage
3 urn:li:mlFeature:(realtime_payment_risk_features,payment_channel) 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,payment_channel)
  • 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: Introduce the unseen digital-wallet channel at the proposed rate and verify that the production encoder's unknown-category behavior remains safe.
  • Expected signals: AUC drop, prediction flip rate, inference error rate
# Operator Column Fraction Value
1 replace_category payment_channel 0.05 digital_wallet

Observed Results

Experiment 1: replace_category

  • Column: payment_channel
  • Duration (how long the mutation and mutated inference took)[ms]: 595.1029000007111
  • Error: n/a
Metric Observed value
baseline_auc 0.8092934782608696
mutated_auc 0.8094565217391305
auc_drop 0.0
prediction_flip_rate 0.0
mean_absolute_probability_shift 0.0006610862130645082
inference_error_rate 0.0
  • Rows selected: 25 / 500
  • Before summary: {"distinct": 3, "missing": 0}
  • After summary: {"distinct": 4, "missing": 0}

Deterministic Decision

  • Status: PASS
  • Summary: All measured impact stayed below configured warning thresholds.

Triggered policy rules

Experiment Metric Observed Rule Result Reason
- none - - - No policy rule triggered

Publication

  • Local JSON: mockup_data/artifacts/runs/20260807T171510Z-10f593.json
  • Local Markdown: mockup_data/artifacts/runs/20260807T171510Z-10f593.md
  • Remote succeeded: yes
  • DataHub tag: Provenance-PASS
  • DataHub document: urn:li:document:shared-95629a75-1b4b-4f2d-9622-6aa3fb8d8b7f
  • Remote error: n/a