Data Scientist

AAR Corp•Wood Dale, IL

7d•$115,000 - $130,000•Onsite

About The Position

The AAR Parts Supply Distribution Analytics Team are developing an AI-driven aviation market intelligence tool designed to transform how aviation parts distribution businesses understand market share, identify growth opportunities, and make strategic decisions. This platform combines advanced analytics, machine learning, agentic AI, and intuitive user experience to provide real-time insights into customers, parts, and markets. The Data Scientist will define, build, and continuously improve the analytical and machine learning models that power the platforms core decision-making capabilities. This role translates complex business problems into scalable, production-ready models that drive market sizing, forecasting, and opportunity identification. The Data Scientist partners closely with product, engineering, and data teams to ensure models are accurate, explainable, and embedded into real-world workflows. Success in this role requires strong technical depth, business intuition, and the ability to operate in ambiguous, data-rich environments. This position is based at our Corporate Headquarters in Wood Dale, IL, with a planned relocation to the Merchandise Mart (Chicago) in early 2027.

Requirements

Strong foundation in statistics, machine learning, and predictive modeling, including regression, classification, clustering, time series, and experiment design e.g., AB testing.
Proficiency in Python and SQL, with experience using common ML libraries such as scikit-learn, pandas, and numpy.
Experience building end-to-end ML workflows, including data preprocessing, feature engineering, model training, evaluation, deployment, and monitoring.
Familiarity with production ML practices, including model versioning, performance optimization, retraining, and monitoring for drift or degradation.
Experience working with large-scale or complex datasets, including handling missing data, inconsistencies, and real-world data limitations.
Ability to communicate model logic, assumptions, and outputs clearly to non-technical stakeholders and cross-functional partners.
Experience collaborating with product managers, data engineers, and software engineers to deliver analytical features in production environments.
Strong problem-solving skills and the ability to operate effectively in ambiguous environments with evolving requirements.
Bachelors in data science, Statistics, Mathematics, Engineering, Computer Science, or a related quantitative field.
3 to 5 plus years of experience in data science, machine learning, or advanced analytics roles in a product or business-driven environment.
Demonstrated experience developing, deploying, and improving predictive models that drive business decision-making.
Experience working on analytics or ML-driven products such as forecasting, optimization, recommendation systems, or market intelligence tools.

Nice To Haves

Master's degree preferred
Experience in designing experiments or statistical validation approaches to evaluate model performance and business outcomes is preferred.
Experience with cloud platforms or distributed data processing such as AWS, Azure, or Spark is preferred but not required.
Experience in aviation, supply chain, or related industries is a plus but not required.

Responsibilities

Design, develop, and own scalable analytical and machine learning models for market sizing, forecasting, opportunity identification, and optimization use cases.
Translate ambiguous business problems into structured modeling approaches, including feature engineering, model selection, and evaluation frameworks.
Design and analyze experiments, statistical tests, and validation methods to measure model quality and business impact.
Deploy and integrate models into production systems in collaboration with data engineering and backend teams, ensuring reliability, scalability, and performance.
Work with large, complex, and imperfect datasets; define data requirements and support robust preprocessing and feature pipelines.
Ensure model outputs are explainable, interpretable, and aligned with business logic to support user trust and adoption.
Monitor, validate, and improve model performance through testing, retraining, versioning, and feedback loops.
Partners with product teams to define analytical features, influence roadmap decisions, and embed model-driven insights into decision-making workflows.