Deep Learning Engineer
Joby Aviation Santa Cruz, CA, USA Competitive Long term > 3 years
About Joby Aviation

It’s getting more costly to get where we want to go. More than a billion people spend over one hour a day in traffic. Transportation emissions are one of the largest sources of pollution. Streets, highways, bridges and tunnels are increasingly expensive to maintain and build.

We aim to solve these problems by delivering safe and affordable air travel to everyone—while advancing the transition to sustainable transportation.

We're a venture-backed startup with offices in Santa Cruz, CA and San Carlos, CA.

Job Description

This position involves developing, testing and deploying algorithms for autonomous flight. Specifically, using deep learning techniques for dynamic obstacles detection and terrain classification. The engineer will play a key role in a small, fast-moving team and have input to conceptual system architectural design and implementation of embedded software to ensure safety of an electric-powered, fly-by-wire aircraft.


  • Design multi-sensor configurations for obstacle detection.
  • Work closely with flight test team to collect large data sets.
  • Post-Process massive, messy, possibly-incomplete data sets to create labelled training and testing datasets.
  • Develop, train, test, and deploy obstacle avoidance algorithms using deep learning techniques.



  • M.S. or PhD in aerospace engineering or computer science, or related field.
  • 3+ years industry experience developing multi-sensor machine learning algorithms with real-world data. 
  • Proficient in C++, Python, and Matlab.
  • Strong foundation of computer vision and deep learning techniques.
  • Basic understanding of version control technologies.
  • Strong work ethic, enthusiasm to learn, and a passion for autonomous, electric, passenger aircrafts.


  • Experience developing parallel algorithms using OpenCL or CUDA.
  • Experience with real-time tracking algorithm - classical or deep learning.
  • Experience with deep learning frameworks such as Pytorch, Caffe, TensorFlow.
  • Familiarity with deep learning approaches for object detection/segmentation such as Deformable-ConvNets, YOLO, Mask-RCNN.
  • Developing systems for probabilistic sensor fusion and data association using cameras, lidar, radar, IMU, GPS, etc.

Please reference you found the job on as thank you to us, this helps us get more companies to post here!