ADAS Engineer

Astemo IndianaFarmington Hills, MI

About The Position

Astemo's Advanced Development Division is hiring a Senior Engineer to own the deployment and optimization of AI/ML workloads, including ADAS perception models, LLMs, and Vision-Language-Action (VLA) models, on embedded automotive SoCs. The engineer will work across multiple areas as priorities evolve and is expected to contribute to both current development needs and emerging software initiatives.

Requirements

  • Knowledge of AD/ADAS systems and automotive hardware platforms.
  • Solid understanding of deep learning fundamentals and the numerical behavior of neural networks.
  • Hands-on experience with major ML frameworks and inference runtimes.
  • Practical experience with model compression techniques and the associated accuracy/performance trade-offs.
  • Working knowledge of embedded SoC architectures and their implications for ML workload performance.
  • Strong programming proficiency in C/C++ (modern C++ preferred) and Python.
  • Familiarity with profiling tools and a structured approach to performance analysis.
  • Exposure to low-level optimization techniques for AI workloads on embedded accelerators.
  • Flexibility and willingness to work across multiple software layers as project needs evolve.
  • Excellent communication and presentation skills, with the ability to influence and persuade stakeholders at all levels of the organization.
  • Ability to work independently with minimal direction.
  • Strong verbal and written communication skills.
  • Experience with PCs and application software, such as MS Office tools.

Responsibilities

  • Deploy and optimize AI/ML models on embedded automotive SoCs to meet performance, memory, and efficiency targets.
  • Apply advanced model optimization techniques while preserving accuracy and intended behavior.
  • Profile inference pipelines and tune workloads (kernels and model graphs) across heterogeneous compute resources for real-time use.
  • Develop orchestration and scheduling approaches for AI workloads across heterogeneous compute resources under real-time and power/thermal constraints.
  • Diagnose deployment-related issues and define validation approaches to evaluate new techniques.
  • Manage and evaluate trade-offs across accuracy, latency, throughput, memory footprint, and energy consumption; produce data-driven recommendations.
  • Collaborate with cross-functional teams to transition advanced work into the production stack.
© 2026 Teal Labs, Inc
Privacy PolicyTerms of Service