
Lead ML/Perception Engineer
May Mobility3 months ago
Base Salary
$235k - $275k/yr
Responsibilities
- Develop software and system requirements with cross-functional teams.
- Lead architectural updates, modularization, and migration strategies for a scalable, maintainable, low-latency perception stack.
- Own scene detection and activation capabilities that classify operating scenarios and trigger appropriate perception behaviors and ODD logic at runtime.
- Improve perception robustness through sensor-fault detection, health monitoring, graceful degradation, and fallback strategies for adverse weather and sensor occlusion.
- Design, implement, test, and deliver production-grade perception features.
- Integrate large-scale multimodal models, including VLMs and LLMs, into the perception stack for semantic scene understanding and reasoning.
- Track and trend perception performance in the field.
- Lead feature design, code reviews, issue diagnosis, and resolution.
- Lead testing to validate features and meet release schedules.
- Lead data, development, and ML pipeline work focused on multimodal data alignment for foundation-model training.
- Set the technical bar, mentor engineers, and make major architectural decisions for real-world autonomous systems.
Requirements
- Minimum 5+ years of industry experience working on real-world robot systems and maintaining high-quality industrial-grade code.
- Master’s degree in Robotics, Computer Science, Computer Engineering, or a field requiring a strong mathematical or engineering foundation.
- Demonstrated experience leading or making major architectural decisions for perception, robotics, or autonomy software stacks.
- Strong programming skills in C, C++, and Python, with software development experience in Linux environments.
- Strong ML/DL development experience with PyTorch or TensorFlow.
- Direct experience developing or fine-tuning large-scale multimodal models, including multi-sensor fusion models, foundation models, and VLM backbones, for real-world perception applications.
- Experience with several real-time areas including computer vision, object detection, classification, segmentation, semantic scene understanding, scene or scenario detection, open-vocabulary detection, degraded-condition perception, multi-target tracking, and sensor fusion.
- Extensive experience deploying features and ML/DL models in real-time systems with high accuracy and low latency.
- Preferred experience with all-weather perception, sensor degradation modeling, fault detection, graceful degradation, ODD monitoring, runtime capability activation, legacy-system re-architecture, synthetic data, reinforcement learning, explainable AI, and ML/DL optimization for resource-constrained real-time products.
- Strong understanding of perception-system architecture, reliability engineering, data pipelines, data balancing, data mining, multimodal learning, contrastive learning, prompt engineering, testing workflows, and production ML deployment.
- Excellent written and verbal communication, attention to detail, rigorous testing practices, and team-leading abilities.
Benefits
- Comprehensive medical, dental, vision, life, and disability coverage, with domestic partner eligibility after one year of cohabitation.
- Health Savings Accounts and Flexible Spending Accounts for healthcare and dependent care.
- Immediately vested employer safe harbor retirement match.
- Paid parental leave with phased return-to-work support.
- Flexible vacation policy and paid company holidays.
- Total Wellness Program resources.
- Standard office working conditions with prolonged sitting, standing, and computer use.
- Moderate travel required, defined as 11%-25%.
Categories
ML EngineeringRobotics
About May Mobility
May Mobility builds autonomous vehicle technology that reasons through the world in real time. Our vehicles have delivered more than half a million rides across 24 deployments in the U.S. and Japan, including driverless operations in three states. We are partnering with Uber, Lyft, and Grab to bring autonomous ride-hail to communities everywhere.