Computer Vision Scientist/Engineer

Job Description

The Role:

As a member of the Autopilot Vision team you will research, design, implement, optimize and deploy models and algorithms that advance the state of the art in autonomous driving. A strong candidate will ideally possess at least one strong expertise in the following areas, and at least a basic familiarity in others.

Job Responsibilities:

  • Train machine learning and deep learning models on a computing cluster to perform visual recognition tasks, such as segmentation and detection
  • Develop state-of-the-art algorithms in one or all of the following areas: deep learning (convolutional neural networks), object detection/classification, tracking, intrinsic/extrinsic camera calibration, visual odometry, structure from motion, multi-sensor fusion, etc.
  • Optimize deep neural networks and the associated preprocessing/postprocessing code to run efficiently on an embedded device
  • Integrate embedded code with the larger Autopilot development team to introduce new features and capabilities to Xpeng’s vehicles.

Job Requirements:

  • Solid understanding of linear algebra, algorithms, machine learning, optimization, numerical methods
  • Excellent C/C++ coding, strong engineering practices, debugging/profiling skills, familiarity with multi-threaded programming
  • Experience with data science tools including Python scripting, numpy, scipy, matplotlib, scikit-learn, jupyter notebooks, bash scripting, Linux environment
  • Experience with at least one main stream deep learning frameworks, including TensorFlow, PyTorch, Caffe(2), MXNet
  • Experience with computer vision or robotics libraries (e.g., OpenCV, PCL, ROS)
  • Experience with CUDA/OpenCL, OpenGL a plus


  • Ph.D. in computer vision or Machine learning
  • Or M.S. with greater than 3 years’ experience in a core computer vision area

Want to apply later?

Type your email address below to receive a reminder

ErrorRequired field

Apply to Job

ErrorRequired field
ErrorRequired field
ErrorRequired field