Perception Engineer – Sensor Fusion

Knightscope•Sunnyvale, CA

3d•$150,000 - $210,000•Onsite

About The Position

Knightscope is seeking a Perception Engineer with a strong AI/ML background to develop and improve the perception stack for our autonomous robotic platforms. This role focuses on multi-sensor data fusion across cameras, LiDAR, radar, and ultrasonic sensors to enable robust environmental understanding and safe robot operation. You will work closely with controls, simulation, and test engineers to deliver perception capabilities that directly impact robot safety, reliability, and performance in real-world deployments. In this role, you will design, develop, and deploy perception algorithms that enable Knightscope robots to understand and navigate complex environments. You will contribute across the perception stack, including 3D object localization, motion estimation, and tracking, while improving the quality of inputs consumed by downstream autonomy systems. This role combines AI/ML model development, multi-sensor fusion, and real-world robotics integration, with a strong focus on performance, robustness, and deployment at scale.

Requirements

S. or Ph.D. in Computer Science, Robotics, Electrical Engineering, Machine Learning, Computer Vision, or related field. B.S. with strong industry experience may be considered
Prior internship or full-time industry experience in robotics, autonomous systems, automotive, aerospace, ADAS, or related fields
Strong background in AI/ML, computer vision, and perception systems
Experience developing multi-sensor fusion algorithms for robotics or autonomous platforms
Experience with 3D perception, object detection, tracking, or scene understanding
Experience working with one or more sensing modalities: camera, LiDAR, radar, ultrasonic, or similar sensors
Strong programming skills in Python and/or C++
Experience developing software in large, distributed development environments
Experience with version control systems (e.g., Git) and collaborative software development workflows (code reviews, branching, merging)
Experience training, evaluating, and deploying machine learning models in production or near-production systems
Ability to work in a hands-on robotics development and testing environment

Nice To Haves

Experience with 3D perception methods (point clouds, voxel representations, BEV-based models, etc.)
Familiarity with modern ML approaches such as deep learning, transformers, or temporal models for perception
Experience with sensor calibration, synchronization, and validation
Familiarity with ROS / ROS2
Experience with real-time inference and embedded or edge deployment constraints
Experience with simulation-based perception validation
Experience supporting downstream systems such as controls, planning, or navigation
Background in functional safety or autonomy system validation

Responsibilities

Design and develop multi-sensor perception algorithms using data from cameras, LiDAR, radar, and ultrasonic sensors
Build and improve sensor fusion pipelines for object detection, classification, tracking, and scene understanding
Develop algorithms for 3D object localization, motion estimation, and tracking in dynamic environments
Implement and optimize perception systems for real-time performance and reliability on robotic platforms
Work with large-scale datasets to train, evaluate, and deploy AI/ML models for perception tasks
Leverage fleet-scale data to evaluate perception performance, identify edge cases, and improve algorithms for robustness, accuracy, and real-world reliability
Define and track perception performance metrics and continuously improve system performance through data-driven iteration
Collaborate closely with controls, simulation, and test teams to ensure perception outputs support safe and reliable robot behavior
Support integration, debugging, and validation of perception systems on real robotic platforms
Contribute to tools, infrastructure, and workflows for perception development and evaluation