Skip to main content
Ctrl+K
FraudTwin documentation is versioned. Check the version selector before copying an API example.

FraudTwin

  • Quickstart
  • Installation and support
  • Choose your path
  • Getting started with FraudTwin
  • Visualization and exploration
    • Core workflows
    • Production ML and reliability
    • Graph analytics
    • Streaming and Kafka reliability
    • Operations and observability
    • Advanced experiments
    • Build and evaluate
    • Integrate and serve
    • Operate and repair
    • Extend and release
    • Core concepts
    • Determinism and reproducibility
    • Data model and lifecycle
    • Point-in-time data and labels
    • Fraud, graphs, and data quality
    • Architecture and trust boundaries
    • Python API reference
    • Configuration parameter reference
    • CLI reference
    • Data contracts
    • Verified capabilities
    • Glossary
    • Troubleshooting
    • Migration guides
    • Compatibility and support policy
    • Development guide
    • References and related work
    • FraudTwin documentation
  • GitHub
  • Quickstart
  • Installation and support
  • Choose your path
  • Getting started with FraudTwin
  • Visualization and exploration
  • Core workflows
  • Production ML and reliability
  • Graph analytics
  • Streaming and Kafka reliability
  • Operations and observability
  • Advanced experiments
  • Build and evaluate
  • Integrate and serve
  • Operate and repair
  • Extend and release
  • Core concepts
  • Determinism and reproducibility
  • Data model and lifecycle
  • Point-in-time data and labels
  • Fraud, graphs, and data quality
  • Architecture and trust boundaries
  • Python API reference
  • Configuration parameter reference
  • CLI reference
  • Data contracts
  • Verified capabilities
  • Glossary
  • Troubleshooting
  • Migration guides
  • Compatibility and support policy
  • Development guide
  • References and related work
  • FraudTwin documentation
  • GitHub

Documentation

  • Getting started

    • Quickstart
    • Installation and support
    • Choose your path

  • Tutorials

    • Getting started with FraudTwin
      • Getting Started with FraudTwin
      • Configure a Simulation
      • Explore Payments and Lifecycle Events
      • Explore Fraud and Delayed Labels
    • Visualization and exploration
      • Visualize payment time, space, and lifecycle behavior
      • Compare Fraud Scenarios, Difficulty, and Camouflage
      • Explore ML-ready distributions, correlation, and embeddings
    • Core workflows
      • From Events to a Trustworthy ML Dataset
      • Investigate and Stress-Test Fraud Scenarios
      • Build a Reproducible Fraud Benchmark
    • Production ML and reliability
      • Build a Simple Fraud Scoring Model
      • Train, evaluate, and track a fraud model
      • Promote and serve a model locally
      • Detect data, domain, and concept shift
      • Promote, reject, and roll back model versions
      • Measure drift by operational segment
    • Graph analytics
      • Investigate fraud with temporal graph exports
      • Build and evaluate point-in-time graph features
    • Streaming and Kafka reliability
      • Publish contracts and inspect Kafka delivery semantics
      • Recover from Kafka outages and duplicate delivery
      • Test Avro compatibility and schema evolution
    • Operations and observability
      • Audit quality faults and build observable projections
      • Materialize a lakehouse snapshot and verify it
      • Resume a scale run from a checkpoint
      • Repair damaged data and replay a bounded interval
      • Persist PostgreSQL rows idempotently
      • Verify Iceberg projections and catalog health
    • Advanced experiments
      • Calibration and counterfactual fraud experiments
      • Campaign dynamics and observable/oracle graph investigation
      • External predictions, temporal backtesting, and feature shift
      • Scale benchmarking, payment reconciliation, and experiment packaging

  • How-to guides

    • Build and evaluate
      • Configuration
      • Data and evaluation workflows
      • ML evaluation methodology
      • Model lifecycle: from simulation to service
      • Data, domain, and concept shift
    • Integrate and serve
      • Production serving reference
      • Integration runbooks
      • Kafka reliability and event-time correctness
      • Spark Structured Streaming
    • Operate and repair
      • Data-quality incidents and replay
      • Graph and benchmark workflows
      • Scale operations
    • Extend and release
      • Extension SDK
      • FraudTwin — Release Readiness & Integration Roadmap
      • Release and benchmark evidence

  • Explanation

    • Core concepts
    • Determinism and reproducibility
    • Data model and lifecycle
    • Point-in-time data and labels
    • Fraud, graphs, and data quality
    • Architecture and trust boundaries

  • Reference

    • Python API reference
      • Typed API cookbook
      • Generation
      • Configuration
      • Data and machine learning
      • Graph
      • Advanced simulation
      • Integrations
      • Benchmarks, quality, and observability
      • Module API reference
    • Configuration parameter reference
    • CLI reference
    • Data contracts
    • Verified capabilities

  • Resources

    • Glossary
    • Troubleshooting
    • Migration guides
    • Compatibility and support policy
    • Development guide
    • References and related work
  • Extend and release

Extend and release#

Use these guides to publish deterministic extensions and prepare evidence for compatibility, release, and benchmark decisions.

  • Extension SDK
  • FraudTwin — Release Readiness & Integration Roadmap
  • Release and benchmark evidence

previous

Scale operations

next

Extension SDK
Edit on GitHub

© Copyright 2026, FraudTwin contributors.

Created using Sphinx 8.2.3.

Built with the PyData Sphinx Theme 0.21.0.