ERS ML Design Optimization Engineer

GMPPU; Concord (NC); On-site; USD 90,000 - 120,000; Full time

Job summary

GMPPU in Concord, NC is seeking an ERS ML Design Optimization Engineer to leverage machine learning for design optimization and performance analysis of ERS systems. The ideal candidate has a strong background in ML and engineering, with expertise in building neural surrogates and optimizing systems for F1.

The role requires a Bachelor's in a relevant field, and a Master's or PhD is preferred. Responsibilities include developing workflows integrated with simulation tools to enhance ERS design while ensuring high fidelity against real-world data.

Qualifications

• Bachelor's in a relevant field; Master's/PhD preferred.
• 3+ years of experience in building neural surrogates.
• Familiarity with hybrid powertrains and multi-physics sim tools.

Responsibilities

• Build neural network surrogates emulating ERS physics.
• Implement optimization loops for multi-objective ERS design.
• Fuse/process petabyte-scale datasets for validation.

Skills

Neural network surrogates

Multi-fidelity datasets handling

GA/BO optimization

Hybrid powertrains knowledge

Domain expertise in e-motors

Education

Bachelor's in CS/EE/Math/Physics

Master's/PhD in ML/scientific computing

Tools

NX

MATLAB

ANSYS

Job description

Formula 1 ERS ML Design Optimization Engineer

Concord, NC, USA

Job Description

GM Performance Power Units (GM PPU) seeks an ERS ML Design Optimization Engineer to join our team in Concord, NC. This role leverages ML for design optimization, simulation acceleration, and performance analysis of ERS systems (MGU-K, CU-K, ES) using telemetry and physics-based data. Focus on surrogate models to reduce sim cycles while meeting FIA constraints.

Key Responsibilities

• Build neural network surrogates (e.g., PINNs, graph nets) emulating ERS physics across thermal, electrical, degradation behaviors.
• Implement tool-agnostic GA/BO optimization loops for multi-objective ERS design (mass/power/reliability).
• Fuse/process petabyte-scale datasets from bench/dyno/track + DiL/HiL/SiL sims for training/validation.
• Conduct sensitivity analysis, uncertainty quantification on ERS parameter spaces.
• Develop ML-accelerated workflows integrated with NX/AVL/MATLAB/ANSYS sim chains.
• Validate models against real duty cycles; iterate for FIA-constrained optima.
• Document optimization pipelines, neural architectures, and results for design reviews.

Required Qualifications

• Bachelor's in CS/EE/Math/Physics; Master's/PhD in ML/scientific computing preferred.
• 3+ years building neural surrogates for engineering sims; GA/BO optimization experience.
• Proficiency handling multi-fidelity datasets (real + DiL/HiL/SiL).
• Familiarity with hybrid powertrains, multi-physics sim tools.
• F1 ERS plant modeling (cell/MGU/ES performance prediction).
• Neural operators/PINNs for PDE surrogates; multi-fidelity BO.
• HPC workflows, data versioning (DVC), containerization.
• Domain expertise in e-motors, batteries, power electronics.

Personal Attributes

• Innovates across model/design/compute trade-offs.
• Communicates complex ML insights to design engineers.
• Rigorous validator of sim fidelity against reality.
• Passionate about F1 performance engineering.