First stable release of Awesome-FCC — a curated Financial Crime Compliance engineering and analytics resource hub.
This release expands the repository from an initial curated list into a broader Financial Crime Compliance (FCC) engineering, analytics and machine learning resource hub.
Highlights
Expanded FCC Coverage
- Improved coverage across AML/CFT, fraud detection, sanctions screening, graph analytics, explainability, model risk and regulatory resources.
- Strengthened vendor-neutral, practitioner-focused FCC resource discovery.
New Sections
- Graph Machine Learning,
- MLOps, Monitoring & Drift Detection,
- Feature Engineering & Feature Store
Added Technical Resources
- Graph analytics & ML: PyTorch Geometric, DGL,
- MLOps & monitoring: MLflow, Evidently AI, whylogs,
- Feature engineering: Featuretools, Feast,
- Data quality: Soda, Pandera,
- Explainability & fairness: LIME, Fairlearn,
- Case investigation tooling: OpenSearch, Apache Superset
Synthetic Data & Simulation
- Added reproducible FCC synthetic data tooling and simulators,
- Included FCC-Synthetic-TM for transaction monitoring experimentation and analytics workflows,
- Added Faker and Mimesis for synthetic KYC/customer generation
Datasets & Benchmarks
Expanded fraud and AML datasets:
- IEEE-CIS Fraud Detection,
- Elliptic Bitcoin AML Dataset,
- PaySim,
- IBM AML Simulated Transactions Dataset,
- Credit Card Fraud Detection Dataset
Regulations & Typologies
Added additional FCC and AML reference sources:
- Wolfsberg Group,
- FinCEN Advisories,
- EU AML Directives / AMLA,
- OFAC guidance,
- FCA Market Abuse resources,
- BIS papers
Improvements
- Expanded README structure and navigation,
- Improved discoverability and practical FCC engineering coverage,
- Minor fixes and formatting improvements
This release provides a stronger foundation for FCC practitioners, data scientists, AML analysts, fraud investigators and engineers working across financial crime domains.