Image Processing Algorithms on mobile GPUs (Mali, Adreno)

Main responsibility: implementing and developing image processing algorithm on advanced mobile GPUs (Mali, Adreno). This includes: - Implementing algorithmic blocks from scratch in OpenCL and OpenGL - Optimizing existing algorithmic blocks, improving run time, reducing data transfer, and improving code efficiency - Deep knowledge of OpenCL/OpenGL drivers on various mobile platforms - Assuming responsibility for all development cycles – design, implementation, debugging, optimization and delivery - Leading GPU SW architecture design - Good knowledge of OpenGL and OpenCL/CUDA. - Excellent C/C++ coding skills. - Knowledge of GPU HW architecture and optimizations. - Experience in designing, implementing and optimizing image processing algorithms on GPU platforms. - Experience in programming for ARM, including NEON intrinsics, assembly-level optimizations - Knowledge of camera architecture / image processing pipeline / 3D stereo algorithms / image registration. - Knowledge of Qualcomm’s Adreno GPU architecture