2027 Summer Intern – Research and Development – Vehicle Aerodynamics and Machine Learning at General Motors in Warren, MI

Job description

Job Description The Role: As a 2027 Summer Intern on the Vehicle Aerodynamics and Machine Learning team, you will work at the intersection of physics-based vehicle simulation, computational fluid dynamics, scientific machine learning, and aerodynamic design optimization. You will support research that uses CFD and AI/ML to improve vehicle aerodynamic performance, accelerate design exploration, and enable more efficient development of next-generation vehicles. The work may include analyzing vehicle flow fields, developing data-driven aerodynamic models, evaluating design concepts, and translating research results into practical engineering tools. This role is intended for an inquisitive, technically motivated student who enjoys applying engineering fundamentals and modern computational methods to complex problems. You will work with experienced researchers and cross-functional engineering partners while developing your skills in vehicle aerodynamics, simulation, machine learning, data analysis, and technical communication. GM’s related research includes high-fidelity CFD, aerodynamic drag and efficiency improvement, physics-based AI models, surrogate modeling, transfer learning, and rapid aerodynamic feedback from vehicle geometry or design data. The Team: The Energy and Propulsion Systems Research team conducts applied research intended to improve vehicle efficiency, performance, range, cost, and sustainability. The team collaborates with researchers, vehicle-development organizations, engineering groups, universities, technology partners, and industry leaders to identify and develop technologies with potential value to GM. You will work in a multidisciplinary environment that may include vehicle aerodynamicists, CFD specialists, data scientists, machine-learning researchers, vehicle designers, optimization experts, and engineers from propulsion, thermal, body, chassis, and vehicle-performance organizations. The internship experience may also include innovation challenges, interactions with industry leaders, cross-functional networking events, technical presentations, and community-service activities. What You’ll Do: Support vehicle aerodynamics research involving external flow, aerodynamic drag, lift, downforce, cooling airflow, and vehicle efficiency. Set up, run, monitor, and post-process CFD simulations for vehicle and component-level aerodynamic studies. Prepare simulation geometry, meshes, boundary conditions, design variables, and analysis workflows. Analyze aerodynamic coefficients, pressure distributions, velocity fields, wake structures, flow features, and other performance metrics. Build and curate datasets from CFD simulations, experimental results, vehicle geometry, and design variables. Develop, train, test, and evaluate machine-learning or surrogate models for aerodynamic quantities and flow-field predictions. Investigate physics-informed, data-driven, reduced-order, graph-based, or other scientific machine-learning approaches relevant to vehicle aerodynamics. Apply data preprocessing, feature engineering, dimensionality reduction, model selection, hyperparameter tuning, and validation techniques. Evaluate model accuracy, robustness, generalization, uncertainty, and limitations across vehicle geometries and operating conditions. Compare ML predictions with CFD, wind-tunnel, experimental, or other trusted engineering results. Support design-of-experiments, surrogate-based optimization, sensitivity studies, and aerodynamic shape exploration. Develop Python-based scripts, tools, and workflows to automate simulation setup, data processing, model training, visualization, and reporting. Help identify opportunities to reduce analysis time and enable earlier aerodynamic feedback during vehicle design. Document assumptions, methods, data sources, model configurations, results, limitations, and recommendations. Create clear plots, visualizations, technical reports, presentations, and demonstrations for engineering and res