Data Scientist, Risk & Fraud

Overview

Whatnot is the largest live shopping platform in North America and Europe to buy, sell, and discover the things you love. We’re re-defining e-commerce by blending community, shopping, and entertainment into a community just for you. As a remote co-located team, we’re inspired by innovation and anchored in our values. With hubs in the US, UK, Germany, Ireland, and Poland, we’re building the future of online marketplaces – together.

From fashion, beauty, and electronics to collectibles like trading cards, comic books, and even live plants, our live auctions have something for everyone. And we’re just getting started! As one of the fastest growing marketplaces, we’re looking for bold, forward-thinking problem solvers across all functional areas. Check out the latest Whatnot updates on our news and engineering blogs and join us as we enable anyone to turn their passion into a business, and bring people together through commerce.

In order to continue this growth, it’s important that Whatnot remains a safe and trusted space to interact and transact. We’re looking for a Data Scientist with expertise in fraud and risk to detect and prevent these threats to our community. You will

• Generate insights & shape direction:
  Translate complex data into actionable recommendations for the Fraud engineering and operations teams.
• Define and own the KPIs that measure the cost of fraud, strategies to prevent it, and impact to users and marketplace performance.
• Analyze the effectiveness of existing methods and partner with product and machine learning engineers to develop better anti-fraud practices.
• Drive experimentation & measurement:
  Partner with product managers, engineers, and operations teams to design, implement, and evaluate feature rollouts to combat bad actors on the platform.
• Define and own the experimentation playbook for Fraud at Whatnot.
• Develop frameworks for causal inference and impact measurement of efforts that are not well-suited to A/B testing.
• Ensure Whatnot’s internal KPIs treat fraudulent actors appropriately in measurement outside of fraud domains.
• Build data products & tools:
  Use our modern data stack to build dashboards, data pipelines, and self-serve tools that empower teams across Whatnot.
• Partner with engineers to improve data accessibility, ensure data quality, and support instrumentation for new product and platform enhancements.
• Lead cross-functional collaboration:
  Advocate for data-driven decision-making and foster a culture of measurement across the trust & risk organization.
• Communicate insights clearly to both technical and non-technical audiences, influencing roadmaps and strategic decisions.
• Bring data support to company-critical investigations to quantify and thwart bad actor tactics, and help generalize outputs to create longer-term protections for different fraud vectors.
• Serve as a thought leader to Trust & Risk leadership, shaping how we build, launch, and iterate on fraud strategy across the platform.

US Based — We offer flexibility to work from home or from one of our global office hubs, and we value in-person time for planning, problem-solving, and connection. Team members in this role must live within commuting distance of our New York, Seattle, Los Angeles, and San Francisco hubs.

Curious about who thrives at Whatnot? We’ve found that low ego, a growth mindset, and leaning into action and high impact goes a long way here. As our next Data Scientist, Risk & Fraud, you bring:

• 5+ years of experience in the Data field, and 3+ years of experience in Data Analytics & Science supporting anti-fraud, risk, trust & safety, or integrity problems.
• Bachelor’s degree in Computer Science, Economics, Statistics, Cybersecurity, or a related field, or equivalent work experience.
• Industry experience with proven ability to apply scientific methods to solve real-world problems on large scale data.

• Advanced SQL skills and experience with modern data warehouses (Snowflake, Big Query, Redshift) and tools like Spark or DBT.
• Proficiency with Python or R for data analysis, modeling, and experimentation.
• E…