Data Science Manager - Fraud

Plaid•New York, NY

10d

About The Position

Fraud Data is the data science and machine learning team within Plaid’s Fraud organization, responsible for using data and ML to improve and scale Plaid’s fraud products. Within Fraud Data, the Customer & Product Intelligence team focuses on understanding product performance, uncovering customer insights, and enabling go-to-market teams with data-driven solutions. The team partners closely with customers and GTM teams on fraud analyses and proofs of concept, turning customer learnings into scalable, reusable product capabilities. We also build the metrics, analytics, and data foundations that measure product health, identify opportunities for improvement, and guide product decisions across Plaid’s Fraud portfolio. As a Data Science Manager, you will lead a team responsible for customer-facing data science and Fraud product analytics. You will set the team's roadmap, develop its data scientists, and remain involved in analytical methods, technical reviews, and customer investigations.

Requirements

Proven experience managing, mentoring, and developing high-performing data scientists.
Deep domain expertise in fraud, risk, or related areas.
Strong experience in product analytics, metric design, and measuring product performance.
Experience partnering directly with customers to deliver data-driven insights and solutions.
Strong technical depth in Python, SQL, statistics, product analytics, and applied modeling.
Demonstrated ability to set technical direction and deliver complex, high-impact initiatives through a team while remaining hands-on.
Excellent communication and cross-functional collaboration skills across Product, Engineering, Machine Learning, GTM, and customer stakeholders.

Nice To Haves

Experience working with graph-based data or systems to identify fraud patterns and improve model performance.
Experience applying causal inference techniques to complex product or risk problems.
Experience using model interpretability techniques across both traditional machine learning and modern model architectures.
Experience with dbt or similar data transformation and analytics engineering tools.

Responsibilities

Set a 6–12-month roadmap with Product, Engineering, and GTM, and assign priorities and responsibilities across the team.
Define product metrics, their underlying data, and reporting and alerting practices; use the results in roadmap and investment decisions.
Establish a repeatable process for customer retrospectives and proofs of concept, including data checks, evaluation methods, and clear recommendations.
Identify fraud signals and product opportunities that recur across customer analyses and work with Product and MLEs to develop them.
Review analytical designs, data models, code, and model evaluations; contribute directly to investigations where your expertise is needed.
Coach data scientists through clear expectations, regular feedback, performance discussions, and growth opportunities.
Use AI-assisted analysis and development tools where useful, and ensure results are properly reviewed before informing customer recommendations or product decisions.
Define how Plaid measures, evaluates, and improves the performance of its Fraud products.
Apply fraud expertise, product analytics, and customer-facing data science to drive end-to-end product and business impact.
Translate customer insights and fraud analyses into scalable product capabilities and opportunities for GTM growth.
Lead and develop a high-performing team while remaining technically hands-on with critical analyses and initiatives.
Raise the bar for product metrics, analytical rigor, and the data foundations that power decision-making across Fraud.