Machine Learning Engineer - Perception

About us:Aeva’s mission is to bring the next wave of perception to a broad range of applications from automated driving to industrial robotics, consumer electronics, consumer health, security, and beyond. Aeva is transforming autonomy with its groundbreaking sensing and perception technology that integrates all key LiDAR components onto a silicon photonics chip in a compact module. Aeva 4D LiDAR sensors uniquely detect instant velocity in addition to 3D position, allowing autonomous devices like vehicles and robots to make more intelligent and safe decisions.  Role Overview:As an ML Engineer on our perception team, you will own the development and deployment of 3D perception models across object detection, semantic segmentation, and lane detection. Your work will directly shape how our autonomous systems perceive and understand the world.What you will be doingTrain and evaluate 3D perception models for object detection, segmentation, and lane detectionImprove model performance across challenging conditions such as long range and sparse point cloudsDeploy perception models to production and own the full lifecycle from research to real-world integrationDiagnose and resolve deployment issues including latency, accuracy degradation, and edge case failures in the fieldMonitor model performance in production and iterate rapidly based on real-world feedbackBuild auto-labeling pipelines using vision-language models to accelerate data annotationWhat will you haveMS or PhD in CS, Robotics, or a related fieldHands-on experience with 3D object detection on LiDAR point cloudsExperience using VLMs for auto-labeling or offline perception tasksStrong Python and PyTorch skillsFamiliarity with large-scale dataset pipelines and annotation workflowsExperience with multi-object tracking or sensor fusion is a plus