Description

You will map and optimize complex embedded C++ codebases for DSP and DNN accelerator platforms. You will own software-hardware partitioning by studying state-of-the-art CV, imaging, and audio algorithms. Your work involves developing and debugging software within real-time, embedded, multiprocessor environments with multiple interfaces. You will propose new algorithms to solve challenging technical problems and maintain complex software environments throughout long project lifecycles.

Responsibilities

• Map and optimize embedded C++ code for DSP and DNN targets.
• Develop software-hardware partitioning strategies using advanced CV and imaging algorithms.
• Debug and implement code in real-time, embedded, multiprocessor systems.
• Propose novel algorithms and creative solutions for complex technical constraints.
• Maintain complex software environments across long-duration project lifecycles.

Required Skills

• 5+ years designing and implementing complex real-time systems using C/C++.
• 5+ years in software development for imaging, computer vision, or signal processing.
• Expert-level C/C++ programming, refactoring, and debugging of heavily threaded code.
• Strong proficiency in algorithms and pattern recognition.
• Experience with 3D graphics and RISC-V architectures.
• Proven ability to solve complex problems and optimize code for efficiency.

Preferred Skills

• MS or PhD in Electrical Engineering or Computer Science.
• Low-level software optimization including instruction-level, loop, vectorization, and caching.
• Experience with OpenCV, Eigen, Halide, or DSP architectures like Tensilica.

Key Skills

Education

Any Graduate