Senior Machine Learning Engineer (Research Scientist) - Fraud

We are the Fraud Data team within Plaid's Fraud organization. We leverage relationships and activity across Plaid’s network to develop machine learning models that help customers detect and prevent fraud while minimizing friction for legitimate users. Our team owns the full model lifecycle, from data processing and experimentation to feature pipelines, model serving, and monitoring. In this research role, you’ll partner closely with Machine Learning Engineers and Data Scientists to uncover new signals, explore and evaluate innovative modeling approaches, and translate promising research into production-ready solutions that strengthen fraud detection across Plaid’s network.

As a Senior Research Scientist, you will lead applied research to develop next-generation fraud detection models across relational graphs, sequential events, images, and video data. You’ll design rigorous experiments and evaluation methodologies that reflect real-world fraud dynamics, while exploring state-of-the-art approaches such as Graph Neural Networks and Transformer-based foundation models. In close partnership with Machine Learning Engineers, you’ll translate promising research into production-ready solutions and communicate your findings internally and externally to advance the technical bar for fraud machine learning at Plaid.

Responsibilities:

- Research and prototype state-of-the-art approaches across graph machine learning, sequential modeling, and multimodal learning to build next-generation fraud detection capabilities.

- Own and execute a research roadmap that translates innovative ideas and prototypes into production solutions with measurable product and customer impact.

- Publish and share applied research while collaborating with a highly skilled, cross-functional team across Data, Product, and Engineering.

- Leverage Plaid’s network-level financial data to uncover insights and develop solutions that help hundreds of millions of consumers achieve greater financial freedom.

Qualifications:

- PhD in Machine Learning, Artificial Intelligence, Computer Science, Statistics, Applied Mathematics, or a closely related field strongly preferred. Candidates without a PhD may be considered with equivalent research experience, such as significant publications, patents, or widely adopted research contributions in relevant fields.

- 2–4+ years of relevant industry or research lab experience, ideally post-PhD, with demonstrated research leadership and a track record of translating innovative research into measurable product or business impact.

- Demonstrated scientific rigor, with strong written and verbal communication skills and the ability to clearly communicate complex research findings.

- Strong proficiency in Python and experience building high-quality research prototypes that can inform or transition into production systems.

Nice-to-Have:

- Experience in fraud detection, security, risk, or abuse prevention.

- Experience with large-scale training, graph systems, and sequential modeling.