Machine Learning Engineer, Frameworks & Performance (New Grad to Principal level)

Qualcomm develops power-efficient on-device and edge AI technologies. The Machine Learning Engineer will develop ML/AI compiler flows and analyze algorithms and workloads to improve performance and inform future hardware and software offerings.

Responsibilities:

• Development and evolution of ML/AI compilers (production and exploratory versions) for efficient mappings of ML/AI algorithms on existing and future HW
• Analysis of ML/AI algorithms and workloads to drive future features in Qualcomm’s ML HW/SW offerings
• Contributing to the development and evolution of ML/AI compilers within Qualcomm
• Defining and implementing algorithms for compiling ML/AI workloads to achieve high performance and low power on Qualcomm HW
• Creating and implementing algorithms that couple PyTorch framework efficiently to Qualcomm ML/AI Compiler flows
• Understanding trends in ML network design, through customer engagements and latest academic research, and how this affects both SW and HW design
• Exploration and analysis of performance/area/power trade-offs for future HW and SW ML algorithms
• Creation of performance-driven simulation components (using C++, Python) for analysis and design of high-performance HW/SW algorithms on future SoCs
• Pre-Silicon prediction of performance for various ML algorithms
• Running, debugging and analyzing performance simulations to suggest enhancements to Qualcomm hardware and software to tackle framework, compute and system memory-related bottlenecks
• Successful applicants will work in cross-site, cross-functional teams