Operations Research & Applied AI Engineer

Roadrunner•Downers Grove, IL

1d•$100,000 - $120,000•Hybrid

About The Position

We are seeking an experienced Operations Research & Applied AI Engineer to build and grow our advanced analytics capabilities. This role combines mathematical optimization with applied artificial intelligence, and it carries responsibility both for maintaining and improving the systems we have today and for identifying, designing, and delivering the ones we do not yet have. The ideal candidate is a hands-on practitioner with genuine depth in optimization and practical experience putting AI applications into production. We care about demonstrated capability rather than credentials, and substantial hands-on experience can stand in place of an advanced degree. This is a builder's role. We expect the person in this seat to look at how the business plans, prices, and moves freight, recognize where a decision is being made by intuition that could be made by a model, and make the case for solving it. Over time, the scope of this position is expected to expand well beyond its starting systems as new opportunities are identified and delivered. This role requires someone who can reason rigorously about model design and solution quality, insist on measurable evaluation before anything reaches production, and then explain the resulting business trade-offs to the leaders who act on them. Experience in the transportation and logistics industry is preferred, as the role involves working with data related to network and capacity planning, service performance, fleet and driver operations, and freight documentation.

Requirements

Bachelor's degree in Operations Research, Industrial Engineering, Applied Mathematics, Computer Science, Statistics, or a related quantitative field (or equivalent experience).
5+ years of experience applying mathematical optimization to production business problems, with demonstrated ability to connect model design to operational outcomes. An advanced degree in a quantitative field may substitute for a portion of this experience.
Strong mixed-integer programming proficiency: ability to formulate, debug, and defend optimization models; interpret LP relaxations, duality, and solver gaps; and diagnose why a model fails to solve within acceptable limits. Hands-on experience with at least one solver such as SCIP, Gurobi, or CPLEX.
Demonstrated ability to design heuristic and metaheuristic methods rather than only apply them, including selecting approaches appropriate to a problem's structure.
Working knowledge of production AI applications built on large language models, including hosted model APIs, structured and schema-constrained output, and failure handling against a non-deterministic service.
Demonstrated evaluation methodology for AI or model output, including ground-truth set construction, accuracy measurement, and making a stochastic pipeline regression-testable.
Strong Python proficiency: typed, tested, packaged production code, with attention to algorithmic performance.
SQL proficiency sufficient to work with relational data sources and to design and maintain output schemas.
Strong analytical and critical thinking skills, with a track record of measuring results honestly and distinguishing improvement from noise.
Demonstrated ability to identify opportunities independently and advocate for them, rather than working solely from an assigned backlog.
Excellent verbal and written communication skills, with the ability to present findings and business trade-offs clearly to both technical and non-technical stakeholders.
Ability to manage multiple priorities and work independently in a fast-paced, results-driven environment.

Nice To Haves

Advanced degree in Operations Research, Industrial Engineering, Applied Mathematics, or a related quantitative field.
Experience in the transportation, logistics, or supply chain industry, particularly network design, load planning, capacity planning, empty repositioning, or driver and crew planning.
Familiarity with Power BI, DAX, SSAS, SSIS, and other data engineering and reporting tools
Familiarity with freight operations concepts such as service standards, hub and terminal operations, cube-versus-weight capacity, and equipment constraints.
Experience with vision or multimodal models applied to document processing.
Experience with bin packing, container loading, or scheduling problems.
Experience with Azure OpenAI or comparable enterprise AI platform services.
Experience packaging and deploying analytical applications for unattended production execution.
Background in statistics, simulation, or forecasting.

Responsibilities

Design, build, maintain, and improve mathematical optimization models supporting network planning, capacity and resource allocation, routing, and related operational decisions.
Formulate and solve large-scale mixed-integer and network flow problems and select appropriate exact or heuristic methods based on problem structure and the time available to solve.
Establish and maintain rigorous solution quality measurement, including derived bounds, so that the value of every model change can be demonstrated rather than asserted.
Translate model parameters, penalties, and constraints into the business trade-offs they represent, and adjust them in partnership with operational stakeholders.
Design and deliver applied AI applications that automate manual processes, extract structured data from unstructured sources, and support operational decision-making.
Move AI solutions from prototype to production, including hosting and scheduling, continuous integration, versioning of prompts and output schemas, structured logging, monitoring, and human review paths for low-confidence output.
Design and maintain evaluation frameworks for AI output, including curated ground-truth sets, accuracy measurement, and regression gates that determine objectively whether a change improved results.
Stay current with developments in applied AI and optimization, and assess which are worth adopting in an operational environment.
Assess business processes across operations, pricing, sales, and finance to identify opportunities where optimization or AI would deliver measurable value.
Build the business case for new initiatives, including expected impact, feasibility, data requirements, and level of effort, and present recommendations to leadership.
Prototype quickly to test whether an approach is viable, and be equally willing to recommend against building something when the evidence does not support it.
Scale successful prototypes into supported production systems with documented design decisions and a defined owner.
Benchmark system output against actual business results to confirm that modeled improvements translate into realized ones.
Distinguish genuine improvement from run-to-run variation, and document negative results as rigorously as positive ones so that approaches are not re-attempted without cause.
Build and maintain the testing and validation infrastructure that keeps analytical systems correct as they change.
Partner with Operations, Pricing, Sales, and Executive Leadership to understand how work is actually performed, and incorporate that operational reality into model design.
Present modeling results, recommendations, and trade-offs clearly to both technical and non-technical audiences.
Service ongoing requests for scenario analysis and decision support, including the projected cost and service impact of proposed operational changes.
Work with security, technology, and business stakeholders to resolve the approvals, data access, and policy decisions required to bring new systems into production use.
Collaborate closely with data engineering and business intelligence partners on shared data dependencies and downstream reporting, and provide guidance and mentorship to team members on analytical and modeling best practices.