Machine Learning Engineer, Data Curation - Autonomous Vehicles

NVIDIA•Redmond, WA

About The Position

NVIDIA is seeking a Machine Learning Engineer to join its Autonomous Vehicles Data Curation team. The role focuses on applying state-of-the-art techniques to find, curate, and evaluate high-quality datasets and ground truth that support autonomous-vehicle products. You will collaborate with researchers and engineers in deep learning and computer vision, translating research into robust, production-ready machine-learning models and taking solutions from initial evaluation and experimentation through deployment. The work supports perception, prediction, planning, and control for self-driving vehicles. It includes defining and collecting datasets, improving scenario coverage and label quality, developing vision-language models and embeddings for scenario search and classification, and building training pipelines and scalable inference systems. NVIDIA describes its AI work as using GPUs to run deep-learning algorithms for applications including robots and self-driving cars.

Requirements

PhD with at least 1 year of relevant experience, or an MS (or equivalent experience) with at least 5 years of relevant experience, in Computer Science, Computer Engineering, or a related technical field.
Proven experience building robust software.
Experience in deep learning or machine learning, with a background in computer vision and/or multimodal learning.
Programming and debugging skills in C++ and/or Python.
Math knowledge.
Good communication and analytical skills, and ability to work with multiple teams in a dynamic environment.
Passion for artificial intelligence applied to robotics and autonomous navigation, willingness to learn, and interest in solving difficult problems.

Nice To Haves

Background applying the latest AI methods to computer vision and autonomous-vehicle problems.
Experience with unsupervised or self-supervised learning and contrastive representation learning.
Experience with vision-language models, agentic search, or automated dataset quality evaluation.

Responsibilities

Stay current with deep-learning research and innovations, and implement and experiment with new ideas.
Collaborate with researchers and engineers to turn research into robust, industrial-strength machine-learning models.
Take approaches from initial evaluation and experimentation through shipping.
Define, collect, and curate training and evaluation datasets, focusing on scenario coverage and label quality.
Develop vision-language models and embeddings for scenario search, classification, and automated data curation.
Build training pipelines and scalable inference systems using technologies such as PyTorch, TensorFlow, TensorRT, Python, and C++.