Director of Data Science

TBK Bank•Dallas, TX

About The Position

At Triumph, our vision is a world where freight transactions are accurate and seamless on the most modern and secure freight transaction network. That’s why we’re looking for passionate, innovative, solutions-oriented people to join our team. We thrive on providing exceptional customer service and we look for team members with an entrepreneurial spirit and a passion to build successful partnerships with our clients. Because at the end of the day our goal is to help our partners businesses run better. Director of Data Science Position Summary We are seeking a Director of Data Science to design, build, and operationalize quantitative models that power both internal and customer-facing intelligence solutions. This role focuses on developing forecasting, optimization, and signal-based models that translate largescale transportation data into trusted insights for carriers, brokers, and shippers. In addition to hands-on technical leadership, this role may include managing data science talent and helping to scale the function over time as organizational needs evolve. The ideal candidate brings strong statistical rigor, experience working with real-world operational data, and a product-oriented mindset for deploying models that influence commercial and operational decisions across transportation networks.

Requirements

A Master's degree in Data Science, Statistics, Mathematics, Economics, Computer Science, or another relevant quantitative discipline is required.
5–7 years of professional experience in data science, applied statistics, or quantitative analytics.
Strong experience with forecasting, predictive modeling, and statistical analysis in applied business contexts.
Demonstrated ability to build models that support decision making, optimization, or market intelligence.
Strong Python and SQL skills and experience working with large, complex datasets.
Experience validating models and monitoring performance in production environments.
Direct experience with model governance frameworks.
Ability to clearly communicate quantitative insights to both technical and non‑technical stakeholders.

Nice To Haves

Experience working with transportation, logistics, supply chain, or network-based data.
Strong expertise in Regression & tree-based models (e.g., XGBoost, Random Forest) Time series forecasting (e.g., SARIMAX, Prophet, TFT) Statistical modeling of skewed distributions (e.g., log-normal, gamma)
2 years in a leadership or people‑management capacity
Deep understanding of freight market dynamics, including the interaction between spot and contract pricing, broker and carrier economics, and the impact of capacity cycles and seasonality on market behavior.
Familiarity with time‑series modeling, signal processing, or index construction.
Experience supporting intelligence, analytics, or data products used by external customers.
Experience working with large-scale datasets in cloud environments and data pipelines (e.g., Snowflake, AWS, Sagemaker)

Responsibilities

Develop and maintain forecasting and predictive models supporting transportation use cases such as pricing, demand forecasting, capacity trends, service performance, and network dynamics.
Build and scale the data science function, including hiring, onboarding, and managing direct reports as business needs evolve.
Design and execute statistical modeling and experimentation, including hypothesis testing, A/B testing, and causal analysis to evaluate market and operational changes.
Build optimization and decision support models that inform routing, capacity allocation, pricing strategy, and operational trade‑offs.
Lead signal development for transportation intelligence products, transforming raw transactional and network data into scalable, reliable indicators and indices.
Establish and lead model validation, performance monitoring, and governance frameworks to ensure stability, accuracy, and trustworthiness of production models.
Partner closely with product, analytics, and engineering teams to translate transportation domain needs into analytically sound, production ready models.
Document methodologies, assumptions, and limitations to support transparency, internal review, and customer facing confidence in intelligence outputs.
Continuously evaluate new data sources, modeling approaches, and techniques relevant to transportation, logistics, and network-based intelligence.