Senior Applied Scientist, Trust & Safety

DAT•Seattle, WA

1d•Hybrid

About The Position

DAT’s Trust and Safety Science team is seeking a Senior Applied Scientist to design and deploy the next generation of risk models and intelligent decision systems that help detect, prevent, and mitigate unsafe, fraudulent, or otherwise harmful behavior across our network. This role sits at the intersection of machine learning, risk decisioning, and product development, with a focus on building systems that protect customers and the marketplace while preserving healthy marketplace activity. You will work on some of the most important trust and safety problems in digital logistics, including onboarding risk, behavioral risk detection, fraud and abuse detection, account integrity, network-graph risk modeling, and continuous monitoring throughout the customer lifecycle. This is a hands-on, end-to-end science role where you will: Conceptualize, propose, implement, and iterate on models and algorithms for fraud detection, risk scoring, and trust and safety decisioning. Build decision engines that learn from feedback and support actions such as step-up verification, review prioritization, and automated access controls. Apply machine learning, graph and network algorithms, anomaly detection, and other quantitative methods to deliver measurable improvements in fraud prevention and operational effectiveness. Take ideas from research to production, ensuring the solutions you build integrate cleanly into operational and product systems. You will be joining at a pivotal point in DAT’s transformation as we automate more of the freight lifecycle and build the safest, most efficient automated marketplace in the freight industry. DAT has also accumulated a uniquely rich set of behavioral, operational, and risk data across its platforms (Convoy Platform, TruckerTools, OutGo, DAT), that enables a strong foundation for behavior-drift modeling, account and identity abuse detection, and broader threat detection systems. A key part of the opportunity is extending Convoy Platform’s industry-leading CARVE product across the broader DAT ecosystem and evolving them into customer-facing risk products for a wider set of DAT customers. This is a deeply technical role focused on building and productionizing high-recall risk models and decision systems for high-stakes compliance and trust workflows, where protecting customers, minimizing missed risk, and making decisions that are measurable, explainable, and operationally defensible all matter. Just as importantly, these systems must act like a scalpel rather than a sledgehammer: in a fair marketplace, we need to target true risk precisely, avoid unnecessary friction for legitimate participants, and make nuanced decisions that balance recall, precision, customer protection, and marketplace health.

Requirements

PhD or MS in Computer Science, Statistics, Applied Mathematics, Operations Research, Engineering, or another quantitative field.
5+ years of experience developing and deploying machine learning, statistical, or decisioning solutions in production environments, with strong proficiency in Python and modern ML tooling and hands-on experience building reliable, production-quality data and model workflows.
Ability to develop algorithmic solutions and decision systems while maintaining explainability, interpretability, and defensibility in high-stakes risk and compliance workflows.
Experience owning a model, service, API, or pipeline end-to-end, including quality, monitoring, iteration, and cross-functional coordination, with strong communication and collaboration skills to work effectively with technical and non-technical partners and bring models into production.
Demonstrated ability to frame ambiguous business problems as scalable automated decision systems and deliver practical solutions with measurable impact.
Experience in one or more of the following areas: fraud detection, trust and safety, risk modeling, anomaly detection, rare-event modeling, identity or abuse detection, graph or network analysis, or related decision systems.

Nice To Haves

You have worked on systems that combine models, heuristics, human review, and operational workflows to make high-stakes decisions.
You have experience with two-sided marketplaces, pricing, financial markets, or economic systems.
You have experience in freight, logistics, transportation technology, or adjacent operational domains.

Responsibilities

Build and productionize fraud, safety, and risk systems for high-recall decisioning, with controls that preserve precision, fairness, and explainability in high-stakes workflows.
Design graph, network-link analysis, entity-resolution, and anomaly-detection algorithms that identify hidden relationships, behavioral drift, account abuse, and emerging threat patterns across users, carriers, digital fingerprints and physical assets.
Develop continuous risk monitoring, alerting, and policy decisioning across onboarding, booking, and load execution, combining ML models, heuristics, feedback loops, and human-in-the-loop review where appropriate.
Move proactively and with urgency against evolving fraud patterns, rapidly iterating on approaches while building scalable, adaptable detection and decisioning systems rather than brittle one-off patches or manual hacks.