Responsibilities:
Work on improving AI frameworks, compiler and CPU performance as well as programmer usability.
Identify areas for improvement in AI framework/compiler toolsets via benchmarking, profiling and code analysis
Collaborate with hardware teams to plan, identify, and contribute support in open source projects for hardware features in Qualcomm silicon
Collaborate with customers to optimize ML models for Snapdragon CPU
Identify areas for improvement in tool usability via interaction with users.
Explore new optimization frameworks for leveraging advanced CPU features.
Design, develop and contribute compiler features to open source projects such as LLVM
Design, develop and upstream optimizations targeting open source ML frameworks.
Minimum qualifications:
Solid understanding of the use and internals of AI frameworks, AI compilers/runtimes and common/emerging AI model architectures, including LLM models.
Proven experience and understanding of neural network layer math and algorithms, including experience with analysis and key kernel optimization techniques
Proficient in working across multiple operating systems(e.g. Linux/Android) , build systems and developer tools
Experience in working with open source communities and open source contribution processes and related tools
Programming in C/C++
Bachelor’s degree in Engineering, Information Systems, Computer Science, or related field.
Preferred qualifications:
Master’s degree or PhD. in Engineering, Information Systems, Computer Science, or related field.
Knowledge and/or experience in compiler frameworks such as GCC or LLVM
Established record of contributions to open source compiler project.
Strong background in computer architecture and compilers