Applied Operations Research Engineer

About Circonomit

Circonomit is pioneering advanced decision infrastructure aimed at industrial enterprises, creating strategic digital twins to optimize complex combinatorial challenges. Originating from RWTH research and supported by a €2.8 million investment led by Vorwerk Ventures, the company empowers the German Mittelstand to navigate market dynamics through precision optimization of production, orders, machinery, and workforce.

Role and Mission

As an Applied Operations Research Engineer, you will bridge sophisticated mathematical optimization and user-friendly software solutions. Collaborating closely with the mathematics and OR teams, you will transform customer-specific production planning challenges—encompassing capacities, costs, constraints, and schedules—into robust, actionable models. These models leverage data from ERP systems and Excel to deliver reliable decisions that plant managers can confidently act upon, even under imperfect data conditions.

Key Responsibilities

• Develop comprehensive customer-centric models by translating industrial planning problems with their associated capacity limits, costs, lead times, and shift schedules into optimization models that guide real-world decisions.
• Design and maintain the abstraction layer that connects software applications with underlying mathematical models, enabling both engineers and customers to extend models independently without direct intervention.
• Ensure the delivery of interpretable and actionable outputs for planners, including transparency about constraint conflicts and the financial implications of potential adjustments.
• Scale optimization models efficiently to handle multiple sites and extended planning horizons while managing simultaneous high-volume customer solves seamlessly.
• Manage your features autonomously from initial development through production deployment and ongoing maintenance, with direct ownership over your deliverables.

Work Environment

You will be part of a small, agile team characterized by minimal bureaucracy, fostering rapid communication and end-to-end project ownership. Continuous feedback and peer reviews are integral, supported by regular in-person collaboration at the Cologne office to expedite problem-solving and enhance team synergy.

Requirements

• Proven experience deploying industrial combinatorial optimization models (MILP, CP, or both) resilient to imperfect data, deadlines, and end-user requirements.
• Expertise in algorithm design with practical deployment, including knowledge of solver capabilities and limits (e.g., CP-SAT, Gurobi), and familiarity with techniques such as warm starts, rolling horizons, relax-and-fix, aggregation, matheuristics, and heuristics when exact methods are unsuitable.
• Experience managing optimization workloads that require timely responses, with solutions for cancellations, timeouts, and parallel processing beyond prototype environments.
• Proficient in writing production-quality Python code emphasizing testing, typing, code reviews, and performance profiling, including deep understanding of numerical challenges interfacing with solvers (scaling, tolerance, integrality, and numeric limits).
• Interest in comprehensive problem understanding including data quality and customer context beyond pure modeling.
• Demonstrated team collaboration capability rather than working in isolation.
• Fluent German (C1 level or above) for team communications and fluent English for code and documentation.
• Resides in or willing to work hybrid within North Rhine-Westphalia regions such as Cologne, Munich, Stuttgart, or Berlin, with adaptability to find the best fitting arrangement.

Preferred Qualifications

• Knowledge of solver internals.
• Experience optimizing numerical or compiled code performance.
• Background in domain-specific languages or compiler development.
• Familiarity with production planning, supply chain, or logistics domains.

Candidate Profile

The ideal candidate is action-oriented, structured, and competitive, driven to deliver practical solutions over academic research freedom. This role requires ownership and the ability to address data challenges directly without waiting for instructions.

Benefits

• Significant impact by building models that directly influence factory planning and industrial decision-making with measurable financial outcomes.
• Full ownership of your work without bureaucratic approval delays.
• Close collaboration with a dedicated math and OR team and the CTO.
• Transparent, rapid feedback culture.
• Attractive compensation including competitive salary and meaningful equity participation (VSOP).
• Choice of preferred hardware and an AI tool budget.
• Membership for sports activities and Deutschland-Ticket for commuting.

Recruitment Process

The hiring sequence consists of an initial 20-minute phone discussion with the CTO, a technical interview, a three-hour hands-on assignment followed by a detailed review session, and final team meetings both online and in-person. The total process spans approximately two to three weeks with prompt feedback after each stage. Candidates are encouraged to request candid conversations with current employees before committing.

Application Guidance

Applicants should omit cover letters and instead share a successful production model they built and reflect on a modeling choice they would reconsider today.

"Hustle the day, analyze at night, reinforce something outstanding."