Supervisors increasingly judge anti-money laundering and counter-terrorist financing regimes on effectiveness rather than paper compliance. For a bank, payment service provider or mobile money operator, that means proving that monitoring scenarios reflect the institution's own risk assessment, that thresholds were set from data rather than habit, that any machine learning in use is explainable and independently validated, and that alerts become intelligence a financial intelligence unit can act on. Analysts who can carry out this work remain scarce.

The Advanced AML / CFT & Financial Crime Analytics Training Course is a ten-day programme for practitioners who already understand customer due diligence and suspicious transaction reporting. Days one to five strengthen the core analytical toolkit: financial crime data and lineage, statistical customer risk scoring and peer-group segmentation, scenario design and calibration, sanctions and PEP screening analytics, and investigation work from triage to a filed report. Days six to ten go further, into transaction graphs and entity resolution, supervised and unsupervised learning, model risk management, and specialist typologies covering trade-based money laundering, virtual assets, terrorist financing and mobile money agent networks.

Each day includes a lab on synthetic data in Python, pandas, SQL or a graph database, so participants practise the methods rather than watch them. The programme closes with an integrated project in which teams design a detection programme for a fictional institution and defend it before a mock supervisory panel.

On completing the ten days, participants will be able to:

• Build an analytics-ready financial crime dataset from core banking, card, mobile money and trade finance sources, with documented lineage and quality checks.
• Design statistical customer risk scores and behavioural peer groups that can be back-tested against past case outcomes.
• Calibrate transaction monitoring scenarios through above-the-line and below-the-line testing and justify each threshold with evidence.
• Evaluate sanctions and PEP screening performance by tuning fuzzy matching and measuring missed and false matches.
• Apply entity resolution and graph analytics to expose mule networks, funnel accounts and layered ownership structures.
• Develop machine learning models that prioritise alerts while keeping their reasoning visible to investigators and supervisors.
• Validate detection rules and models through outcomes analysis, drift monitoring and champion-challenger testing.
• Detect trade-based laundering, virtual asset flows and terrorist financing patterns with specialist analytical tests.

Participants should already work with AML/CFT controls or financial crime data. Typical roles include:

• Money laundering reporting officers and their deputies
• Transaction monitoring and sanctions screening managers
• Financial crime data analysts and data scientists in banks and fintechs
• Financial intelligence unit analysts and strategic analysis teams
• Investigators in bank financial crime units, anti-corruption agencies and police economic crime units
• Compliance staff at mobile money operators, payment service providers and remittance companies
• Central bank and capital markets supervisors who examine AML/CFT systems
• Model risk and internal audit professionals who review monitoring models
• Compliance teams at virtual asset service providers

Participants take back to their institutions:

• A back-tested customer risk scoring model with documented feature weights and segment definitions.
• A threshold tuning workbook recording above-the-line and below-the-line results for each scenario.
• A fuzzy matching test harness for measuring sanctions screening accuracy.
• Reusable Python notebooks for transaction graphs, community detection and mule ring discovery.
• A working alert prioritisation model with SHAP explanations that investigators can read.
• A model validation report template covering data, conceptual soundness, outcomes and drift.
• Analytical checklists for trade invoices, blockchain tracing and terrorist financing indicators.
• A team-built detection programme design and roadmap, reviewed by a mock supervisory panel.

Teaching is analytical and hands-on, and each concept is tested on data the same day. Participants work through:

• Daily coding labs in Python, pandas, scikit-learn, NetworkX and SQL on synthetic banking, mobile money and trade data
• Graph database exercises in Neo4j for link analysis and entity resolution
• Worked case files reconstructed from published FATF typology reports
• Peer critique of tuning decisions, model documentation and STR narratives
• A mock supervisory review in which teams defend their monitoring design
• Pre-built notebooks so that participants with lighter coding experience can still complete every lab

A CPD-accredited Certificate of Completion issued by Vision Reach Global Consultancy is awarded to participants who attend the full ten days, complete the daily labs and contribute to the integrated detection project.

The certificate records the course title, dates and delivery mode, while the notebooks, tuning workbook and validation report built during the programme serve as practical evidence of advanced analytical competence.