Expert Machine Learning Optimization Engineer

VINFAST is a pioneering electric vehicle (EV) company committed to revolutionizing the automotive industry with sustainable and innovative mobility solutions. As a leading player in the EV market, VinFast is dedicated to delivering high-quality, cutting-edge electric vehicles that redefine the driving experience. Our team consists of passionate professionals driven by a shared vision of creating a greener and more sustainable future through innovation, technology, and excellence.

We are seeking a Senior Machine Learning Optimization Engineer to accelerate and optimize the machine learning models powering our self-driving stack. In this role, you will focus on bringing cutting-edge perception , prediction, and planning models to life on real-time automotive hardware. You will design and implement efficient training and inference workflows, ensuring that large-scale models run with the robustness, latency, and efficiency required for safe deployment in autonomous vehicles.

Requirements
Benefits
With respect to all your personal data shared to VinFast in the application and the entire recruitment process of VinFast, by clicking "Apply", submitting your resumé/CV and/or participating in VinFast's recruitment process, you agree that you have read VinFast's Personal Data Protection Policy ("Policy") posted at

or

*, you agree to the Policy and consent for VinFast to process your personal data in accordance with the Policy and the applicable regulations on personal data protection.

To all recruitment agencies: VinFast
does not accept agency resumes. Please do not forward resumes to our careers alias or other VinFast employees. VinFast is not responsible for any fees related to unsolicited resumes.***

• Collaborate with perception , prediction, planning and other teams to optimize models for deployment on embedded automotive processors (e.g., NVIDI A , Qualcom m , TI )

• Profile and analyze CPU/GPU/accelerator performance to identify bottlenecks in runtime, memory usage, and throughput

• Implement and integrate model optimization techniques including quantization, pruning, distillation, mixed-precision training, and efficient neural architectures

• Build and optimize inference pipelines using TensorRT , ONNX, CUDA kernels, and other compilation frameworks to achieve low-latency, real-time performance

• Develop distributed training and scalable optimization pipelines to support large-scale training of self-driving models

• Benchmark and validate optimized models across diverse real-world and simulated driving scenarios, ensuring robustness under edge cases

• Collaborate with system teams to integrate optimized models into the full self-driving stack

• Contribute to CI/CD, build systems, and automated testing pipelines for deployment of optimized ML models

• Ensure compliance with automotive-grade safety, reliability, and performance requirements

• MS c /PhD in Computer Science, Electrical Engineering, Robotics, or a related field with 5 + years of relevant industry experience

• Strong foundation in machine learning, deep learning, and computer vision

• Proficiency in Python and C++

• Hands-on experience with PyTorch or TensorFlow, including large-scale model training and optimization

• Expertise in deploying and optimizing models for embedded or real-time automotive systems using CUDA, TensorRT , ONNX, or equivalent frameworks

• Skilled in profiling and debugging ML code on GPUs and accelerators using profiler tools such as NVIDIA Nsight or similar

• Strong understanding of memory management, parallelization, and performance optimization on embedded processors

• Experience applying software engineering best practices (CI/CD, testing, version control) to ML pipelines

• Excellent problem-solving skills and the ability to thrive in a fast-paced, collaborative team environment

• Nice to have: Prior experience in autonomous driving or ADAS development

• Competitive salary
• Opportunity to collaborate with and learn from industry-leading professionals in the automotive domain.