Optimization Engineer – Commercial

Optimization Engineer – Commercial

The Optimization Engineer – Commercial transforms complex commercial decisions into model-based solutions that improve profitability outcomes and automate manual processes. The position combines operations research modeling skills with software engineering fundamentals to deliver reliable, production-ready workflows. They will iterate on existing optimization logic to enhance model outputs, drive automation in production systems, debug errors in recurring runs, and collaborate with stakeholders on new feature requests.

This role partners closely with the Commercial Data Science team to integrate predictive models into prescriptive processes. The employee will be expected to continuously upskill in the relevant areas and work toward ownership of their assigned systems.

Minimum Requirements Combination of

Education and Experience will be considered. Must be authorized to work in the US as defined by the Immigration Act of 1986. Must pass a Criminal Background Check.

Education:

Bachelor's Degree in Applied Mathematics, Operations Research, Industrial Engineering, Data Science, Statistics, or related field. Years of

Experience:

Minimum two (2) years of experience in a technical environment.

• Proficiency in writing production-quality code in Python, utilizing libraries like Pandas/Num Py and using advanced techniques such as vectorization and parallelization.
• Strong foundation in operations research techniques, including linear/nonlinear/integer programming, network/assignment models, simulation, and/or stochastic optimization.
• Experience building optimization models using Python libraries such as PuLP, Pyomo, OR-Tools, Sci Py, etc.
• Understanding of predictive modeling and forecasting methods such as machine learning, time series, deep learning, reinforcement learning, etc., and their implementation in Python (e.g., scikit-learn, Prophet, PyTorch, Tensor Flow).
• Knowledge of statistical concepts like regression and hypothesis testing for experiment evaluation.
• Competency in querying data using SQL and performing exploratory analysis.
• Ability to clearly communicate with users and stakeholders regarding feature requests, modeling choices, and experiment results through visuals, reports, and demos.
• Demonstrated initiative, curiosity, and an ownership mindset in a fast-paced environment.

Preferred Requirements
• Master's degree or higher in a related field.
• Exposure to airline economics problems such as route planning, capacity allocation, scheduling, pricing, and revenue management.
• Experience with commercial solvers such as Gurobi, CPLEX, FICO Xpress, etc.
• Familiarity with cloud services/model execution environments (e.g., AWS) and version control practices (e.g., Git Hub).
• Familiarity with front-end development in JavaScript and Excel VBA.
• Fluency in the use of generative AI tools to accelerate the software development process, like Git Hub Copilot and Claude Code.

Job Duties
• Formulate airline commercial decision problems into rigorous quantitative models, applying methods from operations research, data science, statistics, and econometrics to increase revenue and decrease costs.
• Build and improve optimization models within Python-based decision support systems by refining assumptions, constraints, objective formulations, and solution strategies to improve performance and stability.
• Design and execute structured evaluations of model features, assumptions, and parameter changes in optimization models using back-testing, controlled experiments, or simulation to assess impact on solution quality, stability, and business outcomes.
• Collaborate with the Commercial Data Science team to develop and incorporate machine learning solutions into relevant decision support tools.
• Extract, transform, and load data from structured/unstructured cloud and non-cloud sources via Python and SQL to create reliable inputs for decision models and recurring workflows.
• Support production systems end-to-end, including troubleshooting data/model issues, diagnosing performance or stability problems, and implementing improvements to robustness, monitoring, and…