Machine Learning Lead

About the role

Coinflow is building a new layer for global commerce by enabling programmable financial infrastructure powered by stablecoins and intelligent automation. The Machine Learning Lead will own the fraud and risk intelligence layer that sits at the center of every transaction decision. This role is about designing systems that learn from live payment behavior and adapt to emerging threats in real time. You will translate messy operational data into structured risk signals that protect revenue and enable growth. The position demands comfort with ambiguity and the ability to turn vague product questions into measurable modeling problems. You will define how the company measures success in fraud, precision, and operational resilience. This is a hands-on leadership role where you write code, review models, and communicate tradeoffs to executives.

Key facts

What you'll do

• Strengthen fraud detection and risk decisioning through feature engineering, model development, and production deployment.
• Own the full model lifecycle, including experimentation, evaluation, monitoring, and iteration.
• Define and track core fraud and risk metrics. These include detection rate, false positive rate, chargeback rate, and dispute win rate.
• Continuously improve core fraud and risk metrics across payment methods and markets.
• Explore transaction and behavioral data to find new fraud signals and emerging attack patterns.
• Partner with Engineering, Product, and Operations to embed fraud intelligence into payment flows and internal tooling.
• Integrate and orchestrate external fraud and risk partners to maximize value from their tools.
• Establish the foundation for ML and data practices across the company, including governance and documentation.
• Help shape Coinflow's long-term fraud, risk, and ML roadmap in alignment with business milestones.
• Translate complex model outputs into clear recommendations for stakeholders with varying technical backgrounds.
• Build experiments that isolate the impact of new features on fraud outcomes and business KPIs.
• Maintain rigorous standards for reproducibility, data quality, and model explainability.
• Mentor and guide junior data scientists and analysts on best practices for modeling in payments.
• Ensure that models remain robust as transaction volumes, merchants, and fraud tactics evolve.

Requirements

• 5+ years in machine learning, applied data science, or production ML roles.
• Demonstrated experience building fraud models in payments. This must include direct exposure to the acquiring side. The candidate should have experience with an acquirer, PSP, or payment facilitator.
• Proven track record of taking ML projects from proof-of-concept to fully deployed, productionized systems.
• Deep familiarity with acquiring-side fraud dynamics. This includes authorization fraud, card-not-present fraud, friendly fraud, chargeback patterns, and merchant risk.
• Strong foundation in ML, statistics, and feature engineering on high-volume financial data.
• Comfortable owning ambiguous problems end-to-end and creating structure where none exists.
• Strong collaborator across Engineering, Product, and Operations.
• Commitment to learning and communicating technical tradeoffs to non-technical stakeholders.

Nice to have

• Experience at an acquirer, ISO, PayFac, or payments infrastructure company.
• Experience developing, managing, and scaling MLOps pipelines and monitoring systems. This includes retraining schedules and real-time performance metrics.
• Experience scoping cloud compute requirements for scalable ML workloads.
• Familiarity with card network rules, dispute/chargeback workflows, and fraud liability frameworks.
• Experience as an early or sole ML hire at a startup.
• Exposure to real-time or near-real-time fraud scoring systems.

Practical notes

The role is based in Chicago, Illinois, and is a full-time position. The compensation structure includes a base salary range from $225,000 to $275,000 USD per year, subject to factors such as experience, education, skills, qualifications, and business needs. Candidates are also eligible for an equity award as part of the total compensation package. There may be performance-based bonuses in addition to base salary and equity. The information provided here reflects the current expectations and requirements detailed in the original job description from the source.