Adreno GPU AI Compiler Performance Specialist

General Summary:

Qualcomms Adreno GPU is the industry-leading mobile graphics solution in todays Android smartphone market and is rapidly expanding into new domains, including the Snapdragon Elite Windows on Arm platform. The Adreno GPU compiler supports a wide variety of GPU programming APIs (OpenGL ES, Vulkan, D3D, OpenCL) and leverages cutting-edge AI-based graphics techniques to enhance visual quality, performance, and user experiences. Join our team to drive innovation in mobile GPU hardware support, advanced compilation techniques, and enhanced gaming and compute experiences on mobile devices and next-generation Windows on Arm systems. The position requires expertise in LLVM-based compiler development and optimization for GPU workloads.

Position Overview

We are seeking a full-time GPU compiler performance engineer to collaborate with application developers, hardware architects, compiler developers, and the graphics research team. This role focuses on analyzing and improving GPU workload performance, enhancing GPU architecture, developing performance modeling methodologies and tools, and leveraging AI-driven approaches for competitive analysis, compiler optimization, and the advancement of AI-based graphics and compilation techniques.

Responsibilities

• Conduct competitive analysis of GPU compiler and performance characteristics across industry leaders (AMD, Intel, Nvidia, ARM), identifying strengths and areas for improvement
• Profile and characterize trending GPU benchmarks and applications (games, HPC, and AI applications), comparing results with competitor platforms
• Utilize external and internal profiling tools, including AI-powered analytics, to analyze performance data and identify bottlenecks
• Apply AI and machine learning methodologies to optimize compiler algorithms and improve GPU workload performance
• Integrate and evaluate AI-based graphics techniques for enhanced rendering, upscaling, denoising, and other visual improvements
• Propose improvements in compilers and GPU architecture, informed by competitive analysis and AI-driven insights
• Recommend application modifications to maximize performance on Qualcomm GPUs, leveraging data-driven and AI-based strategies
• Summarize profiling results and present findings to customers and internal teams, including comparative performance reports
• Collaborate on the development of new AI-based performance modeling and benchmarking tools
• Support performance analysis and optimization for the Snapdragon Elite Windows on Arm platform

Minimum Qualifications:

• Bachelor's degree in Engineering, Information Systems, Computer Science, or related field and 4+ years of Systems Engineering or related work experience
• ORMaster's degree in Engineering, Information Systems, Computer Science, or related field and 3+ years of Systems Engineering or related work experience
• ORPhD in Engineering, Information Systems, Computer Science, or related field and 2+ years of Systems Engineering or related work experience

Qualifications

• BS/MS/PhD degree in Computer Science, Electrical Engineering, or Game Development.
• Experience in compiler development, with exposure to AI-based optimization techniques.
• Hands-on experience with LLVM compiler infrastructure, including developing, optimizing, and debugging LLVM-based GPU compilers.
• Familiarity with LLVM IR, pass development, and integration of AI-based optimization techniques within the LLVM framework.
• Understanding of computer architecture (GPU, memory, data layout, etc.) and performance tradeoffs, including comparative analysis with competitor architectures.
• Proficiency in C/C++ and scripting languages (e.g., Python), with experience in machine learning frameworks.
• Strong communication skills, teamwork spirit, reliability, and self-motivation.

• Experience in graphics programming, OpenCL, or CUDA application development
• Familiarity with performance profiling tools and hardware performance counters for parallel applications on multicore or manycore architectures
• Hands-on experience with machine learning/deep learning tools (scikit-learn, TensorFlow, or similar) for performance analysis and compiler optimization
• Experience with benchmarking and performance tuning for parallel applications, including comparative benchmarking against competitor platforms
• Experience with performance analysis and optimization for Windows on Arm platforms
• Exposure to AI-based graphics techniques such as neural rendering, super-resolution, denoising, and real-time inference for graphics workloads