Lead Perception Engineer

About the job

Lead Perception Engineer

**** An innovative robotics startup is seeking a Lead Perception Engineer to drive the development of advanced computer vision and AI-powered perception systems that are transforming the automotive service industry. This full-time, onsite opportunity focuses on building intelligent vision systems for next-generation robotic tire-changing and wheel-balancing technology. You'll work with cutting-edge technologies including Python, C++, Rust, PyTorch, OpenCV, PCL, Open3D, CUDA, LiDAR, RGB-D cameras, deep learning, and 3D computer vision to help robots accurately perceive, interpret, and interact with their physical environments.

This is an exciting opportunity for a technically strong perception engineer who wants to combine hands-on development with technical leadership. You'll lead the perception engineering team while remaining actively involved in architecture, algorithm development, model training, experimentation, and real-world deployment. One of the biggest advantages of this opportunity is access to proprietary inspection imagery collected directly from real automotive service environments, providing a unique data advantage that traditional robotics companies and public datasets cannot easily replicate.

Required Skills & Experience

• Master's or PhD in Computer Science, Computer Vision, Machine Learning, or a related field, or equivalent technical experience.
• 8–10 years of professional software engineering experience focused on robotics and computer vision applications.
• 2–3+ years of experience in a technical leadership capacity, such as Team Lead, Tech Lead, or technical mentor.
• Proven experience designing, developing, testing, and deploying computer vision and perception systems for robotics or industrial automation.
• Expert-level proficiency in Python, C++, and/or Rust, with experience developing production-quality software for autonomous systems.
• Strong experience with AI/ML frameworks such as PyTorch and computer vision libraries such as OpenCV, PCL, Open3D, or CUDA.
• Experience developing, training, optimizing, and deploying machine learning models.
• Strong understanding of deep learning architectures, including CNNs, ResNet, LSTMs, Transformers, and Vision Transformers (ViT).
• Experience with multimodal AI models, including Vision-Language Models (VLMs) and Vision-Language-Action (VLA) models.
• Strong foundation in robotics, including kinematics, sensor calibration, perception pipeline design, and system architecture.
• Deep understanding of 3D geometry, multi-view geometry, filtering, tracking, and real-time computer vision.
• Experience designing and optimizing high-throughput, low-latency perception pipelines.
• Ability to evaluate sensor technologies and define system-level testing and validation strategies.
• Hands-on experience debugging complex robotic systems and resolving hardware-software integration issues.
• Strong technical communication, collaboration, and mentorship skills.

Desired Skills & Experience

• Experience developing within the ROS 2 ecosystem.
• Experience integrating LiDAR, time-of-flight sensors, stereo cameras, RGB cameras, and RGB-D cameras.
• Experience with 3D reconstruction, object detection, classification, tracking, localization, and pose estimation.
• Familiarity with synthetic data generation and real-world data collection pipelines.
• Experience developing scalable ML training, evaluation, and deployment pipelines.
• Background in industrial automation, autonomous systems, robotics startups, or advanced manufacturing.
• Experience with sensor fusion and multimodal perception architectures.
• Experience deploying computer vision and AI/ML systems in production environments.
• Experience leading cross-functional engineering projects from initial concept through field deployment.

What You Will Be Doing Tech Breakdown

• Programming: Python, C++, Rust
• AI/ML Frameworks: PyTorch
• Computer Vision: OpenCV, PCL, Open3D, CUDA
• Deep Learning: CNNs, ResNet, LSTM, Transformers, ViT
• Multimodal AI: VLMs, VLA Models
• Robotics Middleware: ROS 2
• Sensors: LiDAR, Time-of-Flight, Stereo, RGB/RGB-D Cameras
• 3D Computer Vision: 3D Geometry, Multi-View Geometry, Reconstruction
• Perception: Detection, Classification, Tracking, Localization, Pose Estimation
• ML Engineering: Data Pipelines, Model Training, Evaluation & Deployment
• Robotics: Kinematics, Sensor Calibration, System Integration

Daily Responsibilities

• Lead the perception engineering team through the full development lifecycle, from initial concept and architecture through implementation, testing, and production deployment.
• Define the technical direction and architecture of ATI's perception stack.
• Develop classical computer vision and machine learning solutions for object classification, detection, tracking, localization, pose estimation, and 3D reconstruction.
• Evaluate, select, and integrate sensing technologies, including LiDAR, time-of-flight sensors, stereo cameras, and RGB/RGB-D cameras.
• Build scalable data collection, processing, and evaluation pipelines using real-world and synthetic datasets.
• Develop, train, optimize, and deploy deep learning models, including CNNs, Vision Transformers, temporal models, and multimodal approaches.
• Design high-throughput, low-latency perception pipelines capable of supporting real-time robotic operations.
• Establish performance metrics and lead testing, validation, and sensor calibration to ensure reliable and accurate perception.
• Troubleshoot complex perception challenges and hardware-software integration issues through hands-on debugging in the lab, on the shop floor, and at customer sites.
• Collaborate with mechanical, systems, mechatronics, robotics, software, simulation, and quality engineering teams.
• Establish best practices for software architecture, code quality, testing, and technical documentation.
• Mentor engineers and provide technical guidance to support the growth of the perception team.
• Continuously improve perception algorithms and systems based on real-world performance data and field feedback.

****Applicants must be currently authorized to work in the US on a full-time basis now and in the future.