Lead Machine Learning Engineer - Localization
$235,000–$285,000 year
On-siteAnn Arbor, Michigan, United States
Job Summary
Architect and drive the technical roadmap for a production-grade localization machine learning stack spanning map and sparse landmark-based localization optimized for real-time performance. Lead research, design, training, and validation of advanced neural architectures for object detection, segmentation, and 3D reconstruction to extract localization features. Own the end-to-end data strategy including data curation, auto-labeling, and active learning pipelines to capture long-tail scenarios. Develop robust metrics and evaluation frameworks for feature extraction accuracy and system reliability across diverse Operational Design Domains. Define failure mode criteria ensuring safety case coverage and graceful fallback behavior under sensor or model failure. Evaluate and integrate frontier techniques like multimodal localization and vision/fusion foundation models into production-ready solutions. Drive cross-functional alignment to translate autonomy goals into clear software requirements.
Required Qualifications
- Ph.D. or Master's degree in Computer Science, Electrical Engineering, Robotics, or a related field with a strong mathematical and engineering foundation
- 7+ years of industry experience developing and deploying ML/DL models for computer vision or localization at scale
- Deep expertise in several of the following areas: Computer Vision Foundations: Object detection, classification, segmentation, tracking, depth estimation, 3D reconstruction, and feature detection/description (e.g., SIFT, ORB, SuperPoint)
- Deep expertise in several of the following areas: Vectorized landmark and feature detection networks, BEV-based scene representation, and temporal modeling
- Deep expertise in several of the following areas: Self-supervised/semi-supervised learning, open-vocabulary detection, and vision/fusion Foundation Models
- Deep expertise in several of the following areas: Experience with feature extraction and/or fusion from imagery, LiDAR, and/or radar
- Deep expertise in several of the following areas: Expertise in ML/DL development using PyTorch or TensorFlow, including experience with synthetic data generation, large-scale dataset handling, data curation, and active learning strategies
- Deep expertise in several of the following areas: Strong programming skills in Python and/or C++ with experience in modular software design and Linux-based development
- Deep expertise in several of the following areas: Expertise in ML optimization for real-time products with limited compute, such as quantization and pruning of large transformer models
- Deep expertise in several of the following areas: Proven leadership in developing technical roadmaps, mentoring engineers, and driving measurable improvements in model performance and system reliability
- Moderate: 11%-25% travel required
Desired Qualifications
- 10+ years of experience in ML/DL for autonomous driving or ADAS systems
- Experience utilizing Vision-Language Models (VLMs) and/or Foundation Models for auto-labeling and long-tail (edge-case) detection
- Working knowledge of localization and state estimation concepts (e.g., SLAM, sensor fusion)
- A proven record of inventions and/or publication record at top-tier conferences (e.g., CVPR, NeurIPS, ICCV, ECCV, ICLR)
Hiring someone like this?
Get your role in front of qualified candidates on Sorce.