About the job
About impact.com
impact.com stands at the forefront of commerce partnership marketing, reshaping how businesses flourish by facilitating the discovery, management, and expansion of partnerships throughout the customer journey. From affiliates and influencers to content creators, brand advocates, and customer supporters, impact.com empowers brands to drive authentic, performance-driven growth through meaningful collaborations. Our award-winning products, Performance (affiliate), Creator (influencer), and Advocate (customer referral), integrate all partner types into one cohesive platform. As consumers increasingly seek recommendations from trusted sources, impact.com ensures brands are visible where it counts. Today, over 5,000 global brands, including Walmart, Uber, Shopify, Lenovo, L’Oréal, and Fanatics, depend on impact.com to fuel more than 225,000 partnerships that yield measurable business outcomes.
About the Role
We are on the lookout for a Principal Data Scientist with expertise in Fraud and Risk to join our dynamic Data Science team in Cape Town. In this pivotal role, you will lead efforts to safeguard our affiliate marketing ecosystem by researching, developing, and implementing machine learning models aimed at detecting and preventing fraud related to attribution, lead quality, and partner compliance. You will tackle high-stakes challenges that encompass traditional fraud patterns and emerging threats, from attribution manipulation to browser extension misuse, while creating scalable production systems. This position offers a unique blend of rigorous analytical work with substantial business impact in a rapidly evolving, adversarial environment.
Core Responsibilities
Research & Model Development
- Engage in research and development initiatives focused on fraud detection and risk management within the digital advertising ecosystem, addressing issues such as attribution fraud, lead fraud, click injection, browser extension abuse (e.g., coupon hijacking), brand safety violations, and the verification of creator authenticity.
- Design, prototype, and validate machine learning models and rule-based systems for fraud detection, partner risk assessment, compliance monitoring, and trust & safety workflows.
- Explore and apply graph-based fraud detection methodologies (including community detection, link analysis, and behavioral clustering) and investigate graph database applications for modeling the relationships among users, devices, transactions, and partners to unveil coordinated fraud rings and unusual network patterns.
- Proactively monitor emerging fraud patterns through continuous research...
