Remote Fraud ML Engineering Manager

Rozgoo; United States; Remote; USD 178,000 - 228,000; Full time

Benefits offered by this job

Remote-friendly

Equity rewards

Medical coverage

Job summary

Affirm’s Fraud Machine Learning team seeks a manager to guide a group of ML engineers in building and refining fraud detection models across the full ML lifecycle. You will shape modeling strategy, lead end-to-end development, and collaborate with product, risk, and platform teams to ensure robust, production-ready solutions.

You will advocate advanced techniques such as representation learning and transformer-based approaches while maintaining strong engineering fundamentals and ownership.

Qualifications

• Bachelor’s degree in a technical field (or equivalent practical experience)
• 8+ years of industry experience with 3+ years managing engineers
• Experience with modern ML approaches: representation learning, deep learning, transformer models, and gradient-boosted trees

Responsibilities

• Set technical and modeling strategy for fraud detection across the ML lifecycle
• Lead a team of ML engineers from experimentation to production
• Drive evolution of modeling approaches including transformer-based methods
• Partner with product, analytics, risk, and eng to deploy models in decisioning systems
• Develop talent by coaching engineers and promoting a high-performing culture

Skills

Leadership

ML modeling

Transformer models

Representation learning

Production ML

Cross-functional collaboration

Scalable systems

Education

Bachelor’s degree in a technical field

Equivalent practical experience

Tools

Python

ML platforms

Job description

Affirm’s Fraud Machine Learning team seeks a manager to guide a group of ML engineers in building and refining fraud detection models across the full ML lifecycle. You will shape modeling strategy, lead end-to-end development, and collaborate with product, risk, and platform teams to ensure robust, production-ready solutions.

You will advocate advanced techniques such as representation learning and transformer-based approaches while maintaining strong engineering fundamentals and ownership.