Senior Embedded Engineer Camera Vision at Samsara

About the role:

The role reports to the safety embedded applications team. The engineer will be working closely with the machine learning team members in integrating the ML models into the firmware and optimizing, productizing the edge ML models. 

In this role, you will: 

• Be responsible for developing and integrating camera offerings for Samsara’s dash camera and fleet management system.
• Integrate core camera imaging features and audio features, by working closely with the SoC vendors and ODM partners. 
• Be responsible for core camera performance tuning related to image quality, audio quality and machine vision featureset.
• Work closely with the AI/ML team and other firmware teams to enable safety features for Samsara dash cameras, Primarily working on the camera integration  and productization of the AI/ML models for our cameras to improve fleet safety.
• Champion, role model, and embed Samsara’s cultural principles (Focus on Customer Success, Build for the Long Term, Adopt a Growth Mindset, Be Inclusive, Win as a Team) as we scale globally and across new offices.

Minimum requirements for the role:

• 5 to 8 years of experience in embedded software development on complex SoC using Linux operating systems or similar. 
• Proficient in C or C++ programming language with strong debugging skills on embedded systems. 
• Understanding of the camera systems operation of image sensors, 3A algorithms and other image processing algorithms. 
• 1 to 2 years of experience in embedded computer vision frameworks and integration of AI/ML models into embedded systems.
• Experience in ADAS embedded system deployment involving cameras. 
• Awareness of camera image quality tuning tools & fundamentals. Awareness of audio/video encoding standards.

An ideal candidate also has:

• Additional skills in complex audio video multimedia systems using frameworks like gstreamer and NNstreamer. 
• Experience in the ADAS systems and its associated technical complexity in the field of embedded systems.
• Understands the challenges in productizing computer vision models in edge computing starting with datasets, model conversions to run on edge, integration of the models and deployment to production.
• Familiarity with golang and edge to cloud interfaces using protocols such as gRPC, websockets.
• Familiarity with edge ML frameworks like Tflite, SNPE, TVM etc.
• Familiarity with OpenCL, OpenGL, Hexagon DSP. 
• Familiarity with distributed computing across DSP, GPU, NPU, CPU.