Machine Learning Engineer, Fraud at whatnot

This role is ideal for a mid-level Engineer (2–6 years of experience) who thrives at the intersection of backend engineering and data science. You should have a

Work type: onsite

Location: San Francisco, CA | Seattle, WA | New York, NY | Los Angeles, CA

Salary: $245,000 – $345,000/yr

Type: Full-time

Summary

This role is ideal for a mid-level Engineer (2–6 years of experience) who thrives at the intersection of backend engineering and data science. You should have a proven track record of deploying production-grade models, specifically within risk, fraud, or trust and safety domains. The team is looking for someone who can transition comfortably between building scalable data pipelines and conducting deep adversarial analysis to catch bad actors in real-time. The compensation is highly competitive, starting at $245,000 plus equity and a 4% 401k match. While the role is "on-site," Whatnot operates as a remote-co-located team, offering a "best of both worlds" approach: hefty work-from-home stipends and flexibility, paired with local hubs in SF, NYC, LA, and Seattle for in-person collaboration. Unique perks include a monthly budget to "dogfood" the app by buying and selling collectibles. **You might be a good fit if you:** * Have experience building intelligent user graphs or anomaly detection systems to stop fraud. * Are proficient in Python and ML frameworks like PyTorch or LightGBM. * Can manage the full lifecycle of a model, from SQL/Spark ETLs to real-time inference and drift monitoring. * Want to work in a high-growth, fast-paced marketplace environment.

Job Description

## 🚀 Join the Future of Commerce with Whatnot!

Whatnot is the largest livestream shopping platform in North America and Europe to buy, sell, and discover the things you love. Whether it's trading cards, fashion, electronics, or live plants, our sellers are building real businesses across hundreds of categories. We're building live commerce at a scale that's never been done in the West, and there's no playbook to copy. The people here are shaping how an entirely new industry develops.

As a remote co-located team, we're inspired by our [values](https://www.whatnot.com/careers) and anchored in hubs across the US, UK, Ireland, Poland, Germany, and Australia. We move fast, stay close to our users, and focus on the work that drives the most impact.

We're one of the [fastest growing marketplaces](https://a16z.com/marketplace-100/) and were recently named the [#1 Best Startup Employer in America](http://google.com/search?q=%231+forbes+startup+employer&oq=%231+forbes+startup+employer&gs_lcrp=EgZjaHJvbWUyBggAEEUYOTIGCAEQRRg9MgYIAhBFGD0yBggDEEUYQDIGCAQQRRhAMgYIBRBFGEDSAQg1NzM0ajBqMagCALACAA&sourceid=chrome&ie=UTF-8) by Forbes. Check out the latest Whatnot updates on our [news](https://blog.teamwhatnot.com/) and [engineering blogs](https://medium.com/whatnot-engineering) and join us as we enable anyone to turn their passion into a business and bring people together through commerce.

## 💻 Role

• Design, train, and deploy both traditional ML and LLM-powered models to detect fraudulent behaviors across users, payments, and marketplace interactions.

• Lead the end-to-end architecture of fraud detection, prevention, and intervention systems — balancing platform security with a seamless user experience.

• Build intelligent user graphs to model behavioral patterns, collusion networks, and account connectivity.

• Develop scalable data pipelines and real-time inference systems supporting high-volume, low-latency ML workloads.

• Conduct deep behavioral and adversarial data analysis to uncover fraud trends and continuously improve detection accuracy.

• Partner cross-functionally with Trust & Safety, Payments, and Infrastructure teams to develop features, labels, and model evaluation pipelines.

• Implement model monitoring and drift detection systems to ensure reliability and responsiveness.

• Contribute to fraud risk orchestration, combining rules, models, and heuristics for decision automation.

• Define and track key metrics and dashboards for fraud detection effectiveness (e.g., precision, recall, false-positive rate, latency).

• Stay ahead of emerging fraud tactics and continuously translate insights into adaptive, production-ready systems.

## 👋 You

People who do well at Whatnot tend to be comfortable figuring things out as they go, biased toward action, and genuinely curious about what they're building. They care more about outcomes than credit and stay close to the product and the people using it.

• Bachelor’s degree in Computer Science, a related field, or equivalent work experience.

• 2–6 years of experience in machine learning or software engineering, ideally in risk, fraud, or trust & safety domains.

• Strong proficiency in Python and ML libraries (e.g., scikit-learn, PyTorch, LightGBM).

• Solid backend development skills and experience deploying ML models to production (batch or real-time).

• Experience in data analysis and ETL (SQL, Spark, DBT) for data pipeline building.

• Familiarity with fraud detection techniques such as chargeback prediction, anomaly detection, or graph-based modeling.

• Experience with data orchestration frameworks (Dagster, Kubeflow) and feature store design.

• Ability to translate business risk into measurable ML solutions and collaborate across diverse

For US-based applicants:$245,000 - $345,000/year + benefits + stock options

The salary range may be inclusive of several levels that would be applicable to the position. Final salary will be based on a number of factors including, level, relevant prior experience, skills and expertise. This range is only inclusive of base salary, not benefits (more details below) or equity in the form of stock options.

🎁 Benefits

• Generous Holiday and Time off Policy

• Health Insurance options including Medical, Dental, Vision

• Work From Home Support

• Home office setup allowance

• Monthly allowance for cell phone and internet

• Care benefits

• Monthly allowance for wellness

• Annual allowance towards Childcare

• Lifetime benefit for family planning, such as adoption or fertility expenses

• Retirement; 401k offering for Traditional and Roth accounts in the US (employer match up to 4% of base salary) and Pension plans internationally

• Monthly allowance to dogfood the app

• All Whatnauts are expected to develop a deep understanding of our product. We're passionate about building the best user experience, and all employees are expected to use Whatnot as both a buyer and a seller as part of their job (our dogfooding budget makes this fun and easy!).

• Parental Leave

• 16 weeks of paid parental leave + one month gradual return to work *company leave allowances run concurrently with country leave requirements which take precedence.

Whatnot is proud to be an Equal Opportunity Employer. We value diversity, and we do not discriminate on the basis of race, religion, color, national origin, gender, sexual orientation, age, marital status, veteran status, parental status, disability status, or any other status protected by local law. We believe that our work is better and our company culture is improved when we encourage, support, and respect the different skills and experiences represented within our workforce.

View this job on nocollar jobs