Senior Data Scientist / ML Engineer (Credit risk) - Long term contract
Working hours: Mon - Fri, 8:30am - 5:30pm (1-hour lunch break included) - GMT+7 (Vietnam time)
Location: Remote - Vietnam
Overview: You will build the analytical and detection core of one of our key risk infrastructure products, serving financial institutions across Southeast Asia. You'll turn raw financial data into enriched features, credit risk signals, and fraud detection models and you'll ship those models into production yourself. This is a deeply technical, build-oriented role. You'll design and validate models, write the backend code that runs them at scale, and own the data pipelines behind them. You should think like a quant or credit modeler & build like a backend engineer.
Design, build, and validate credit risk and fraud detection models - income estimation, cash flow and affordability analysis, transaction categorization, anomaly and fraud pattern detection.
Engineer enrichment and feature pipelines that transform raw financial data into structured, decision-ready signals for underwriting.
Write production backend code (Python and/or equivalent) to deploy models and enrichment logic as services and APIs - not just prototypes, but code that runs in production.
Develop and tune fraud detection logic: rules engines, statistical anomaly detection, and ML-based approaches; define fraud typologies and translate them into detectable features.
Own model performance end to end - define and monitor detection rates, false positive rates, precision/recall, PSI/stability, and coverage; retrain and recalibrate as data shifts.
Build backtesting, validation, and monitoring frameworks to ensure model quality, explainability, and regulatory defensibility.
Partner with engineering squads on data architecture, pipeline scalability, and deployment; partner with product and BD to align models with real underwriting workflows.
Document methodology and model governance to a standard financial institutions and auditors can review.