Development Engineer IV - Engineering Data Scientist and Digital Twin Specialist

Join the world's largest HVAC company, named by Forbes as one of America's Best-In-State Employers 2025!

Daikin Applied is seeking an Engineering Data Scientist & Digital Twin Specialist with a strong focus on Reduced Order Modeling (ROM). In this role, you will bridge the gap between high-fidelity 3D physically-based (FEA/CFD), 1D system performance simulations, lab, and real-time operational data. You will build, validate, and deploy fast-running surrogate models and hybrid digital twins that power predictive maintenance, real-time edge analytics, and automated design optimization for our physical assets and systems.

Come be a part of an exciting journey at Daikin Applied, where innovation and excellence drive our every endeavor!

Hybrid
- Plymouth, MN

• Reduced Order Modeling (ROM): Develop, calibrate, and validate ROMs from complex 3D/multiphysics simulations (e.g., thermal, structural, fluid dynamics) to accelerate computation speeds by orders of magnitude without losing fidelity.
• Hybrid Digital Twin Development: Design and implement hybrid digital twins that combine first-principles physical models with machine learning/AI (physics-informed neural networks, surrogate modeling) to mirror real-world asset behavior.
• Data Integration & Pipelines: Ingest, clean, and utilize high-frequency time-series telemetry and IoT sensor data from physical machinery/assets to continuously update and retrain digital models.
• Deployment & Scaling: Package and deploy ROMs into production environments, cloud platforms, or real-time edge devices using platforms like Ansys Twin Builder, Siemens Simcenter, or custom Python/C++ frameworks.
• Cross-Functional Collaboration: Work tightly with domain engineers, software developers, and data engineers to integrate digital twin frameworks into broader enterprise architectures and PLM.
• Model Validation: Conduct rigorous regression testing, scenario analysis, and test-data correlation to ensure numerical stability and accuracy against physical counterparts.

• Master's or Ph.D. in Mechanical Engineering, Aerospace Engineering, Computer Science, Applied Mathematics, Data Science, or a related technical discipline
• 6+ years of industry/research experience in applied machine learning, scientific computing, or physics-based simulation
• Proven track record of building and deploying Reduced Order Models (ROMs) (e.g., Proper Orthogonal Decomposition (POD), Dynamic Mode Decomposition (DMD), or machine learning surrogates like Gaussian Processes and neural networks)
• Advanced proficiency in Python (Num Py, PyTorch/Tensor Flow, Scikit-learn) and/or C++
• Familiarity with engineering simulation software suites (e.g., Ansys Twin Builder, Siemens Simcenter, MATLAB/Simulink, or OpenFOAM/FEA tools)
• Experience with time-series databases, IoT data streams (MQTT, OPC UA), and containerization (Docker, Kubernetes) for model deployment
• Strong understanding of physical principles (dynamics, thermodynamics, heat transfer, structures, or fluid mechanics) alongside statistical modeling and machine learning
• Strong communication and presentation skills, with the ability to clearly convey technical concepts to both technical and non-technical audiences
• Demonstrated ability to lead technical project teams and mentor engineers
• Knowledge of systems engineering and architecture principles
• Demonstrated ability to work independently and drive collaboration in a cross-functional, globally distributed environment
• Understanding of model reuse, simulation governance, and lifecycle management concepts
• Track record of leading cross-disciplinary simulation initiatives or shaping organizational modeling strategy

• Experience with MiL and HiL simulation workflows
• Experience with machine learning, data analytics, or AI-assisted modeling and automation
• Background in experimental data acquisition and validation of simulation models using test data
• Experience with physics-informed machine learning (PINMs) or geometric deep learning
• Exposure to industrial IoT platforms or…