Lead Data Scientist - Fraud & Risk, ML for Trust & Safety

ATS Client; Cape Town; On-site; ZAR 1,000,000 - 1,400,000; Full time

Benefits offered by this job

Flexible Working

Health and Wellness

RSUs

Coursera subscription

Parental leave

Technology stipend

Job summary

impact.com in Cape Town seeks a Lead Data Scientist focused on Fraud and Risk to protect our affiliate ecosystem. You will research, develop, and deploy ML models that detect fraud across attribution, lead quality, and partner compliance, while building scalable production systems.

You will work with Product, Compliance, and Trust teams to translate insights into business actions, monitor performance, and continuously improve models in a dynamic environment.

Qualifications

• Experience 5+ years in data science with fraud/risk in production environments.
• Fraud and risk domain expertise in building and deploying models.
• Strong Python and SQL; ML libraries and feature engineering.
• Bachelor's in a quantitative field; Master's/ PhD preferred.

Responsibilities

• Research and develop ML models for fraud detection and risk scoring.
• Deploy end-to-end fraud/risk solutions from data to monitoring.
• Collaborate with Product, Compliance, and Trust teams to drive impact.
• Analyze fraud trends and communicate actionable insights to leadership.

Skills

Python

SQL

Fraud detection

Risk modeling

Production ML workflows

Statistical analysis

Communication

Graph analytics

Education

Bachelor's in a quantitative field

Master's/PhD preferred

Tools

scikit-learn

XGBoost

LightGBM

Graph databases

Job description

impact.com in Cape Town seeks a Lead Data Scientist focused on Fraud and Risk to protect our affiliate ecosystem. You will research, develop, and deploy ML models that detect fraud across attribution, lead quality, and partner compliance, while building scalable production systems.

You will work with Product, Compliance, and Trust teams to translate insights into business actions, monitor performance, and continuously improve models in a dynamic environment.