Developer for the Sensor Data Delivery Team

The foundation of autonomous driving is the flawless operation of sensors such as cameras, lidars, and radars. Our team is responsible for integrating sensors into the autopilot system and creating a reliable and efficient pipeline for sensor data delivery. We are constantly improving the system and testing new device models. We are looking for a developer who will help integrate new hardware into our autonomous platform.

What tasks await you

Development and support of drivers for various sensors New, more advanced sensors are emerging, delivering an ever-increasing data stream, and new types of sensors are appearing. It is necessary to implement software for interacting with them — both for "on the desk" experiments and production-level within the platform, ensuring stable delivery of sensor data with minimal latency.

Optimization of sensor data delivery You will be designing architecture, implementing preprocessing of raw data from sensors and its conversion into unified formats and protocols for further processing by perception algorithms. You will implement synchronization of data from various sensors to build a comprehensive picture of the environment.

Optimization of resource consumption New generations of sensors deliver an order of magnitude larger volume of data, the number of sensors installed on the autonomous platform is increasing, and data processing latencies must decrease. It is necessary to optimize the architecture and algorithms of the pipeline to provide consumers with a complete set of information about the surrounding world.

We expect you to

• Be proficient in C++
• Understand the architecture of systems based on Linux
• Understand the principles of computer networks, have worked with main network protocols
• Be able to work in a team

Will be a plus

• Have worked with Ethernet networks
• Are familiar with digital interfaces CAN, GMSL
• Have used the ROS or ROS 2 framework
• Understand the specifics of using Real-Time Operating Systems and developing multithreaded applications