Senior Robotics Perception Engineer
Senior Robotics Perception Engineer at Pave Robotics — San Francisco, CA, US
- Company: Pave Robotics
- Location: San Francisco, CA, US
- Employment type: FULL_TIME
- Salary: USD 150000–200000 / year
- Posted: 2026-07-17
About this role
This is a full-time, in-office opportunity to work at the cutting edge of robotics in unstructured real-world environments. If you thrive in a fast-paced startup environment and want to push the boundaries of robotics engineering, this role is for you.
Key Responsibilities
Build out early-stage robot to work in real-world unstructured road environments
Create motion and path planning algorithms for high-speed crack sealing
Develop robust perception algorithms for real-time crack detection
Implement sensor fusion algorithms for reliable localization and state estimation (encoders, IMU, cameras, etc.)
Work with a team to deploy AI solutions in real-world applications, focusing on improving model performance
Develop and implement novel approaches to model fine-tuning, optimization, and evaluation
Optimize robot system-wide to hit engineering requirements across accuracy, speed, and reliability
Qualifications
Bachelor’s, Master’s, or PhD in Computer Science, Math, Robotics, or relevant field
3-5 years at an early stage startup
Experience with ROS/ROS2, Python, C++
Expertise in one or more of the following core robotics competencies:
Localization/mapping
Perception
Motion/path planning
Motion control
Simulation + sim-to-real
Experience with deep learning frameworks such as PyTorch or TensorFlow
Experience with applied machine learning in a robotics context (image classification/segmentation preferred)
Knowledge and experience with robot simulation (Gazebo, IsaacSim, etc.)
Experience with containerization platforms such as Docker
Knowledge and experience working with Jetson, Arduino
Experience working cross-functionally across hardware and software environments
Ability to rapidly prototype solutions, perform field testing, and iterate on software design
Ability to quickly diagnose complex problems system-wide across multiple domains of robotics
Portfolios highly regarded