AI TLM Performance Modeling

Teknik, data och digitalt · Data, AI och analys · Maskininlärning · Mjukvaruutveckling · Data engineering

I korthet

You will be involved in building performance modeling platforms and tools for Qualcomm’s SoC hardware, specifically for AI/cloud data server applications, analyzing and validating LLMs and AI workloads. This role involves working on performance modeling of Qualcomm’s AI hardware platforms to find optimal architecture and tradeoffs, and optimizing LLM frameworks and AI workloads.

Ansvarsområden

Build performance modeling platforms and tools for Qualcomm’s SoC hardware targeting AI /cloud data server applications.
Analyze and validate LLMs and AI workloads for cloud data server applications.
Work in a dynamic research environment with a multi-disciplinary team of hardware and system engineers.
Optimize LLM frameworks and AI workloads.
Design, develop & test performance modeling platforms and tools for analyzing and optimizing AI workloads to improve performance and power efficiency on all devices.

Krav

Programming in Python, C/C++, SystemC, TLM2.0.
Experience in SoC architecture, DDR memory controller, NOC design/modeling.
Experience in AI workloads and architectures including LLMs and ML networks.
Experience in architecture for machine learning hardware accelerators.
Experience with TensorFlow, Pytorch.
Hands-on experience with performance measuring, analysis of ML networks.
Familiarity of Computer architecture, DSP architecture and hardware(RTL) design/modeling/verification.
Experience with compiler frameworks such as LLVM or GCC.
Bachelor's degree in Computer Science, Engineering, Information Systems, or related field and 4+ years of relevant work experience.
OR Master's degree and 3+ years of relevant work experience.
OR PhD and 2+ years of relevant work experience.

Förmåner

Competitive annual discretionary bonus program.
Opportunity for annual RSU grants.
Highly competitive benefits package designed to support your success at work, at home, and at play.

Färdigheter

PythonC++SystemCTLM2.0SoC architectureDDR memory controllerNOC designAI workloadsLLMsML networksTensorFlowPytorchComputer architectureDSP architectureRTL designLLVMGCC

#AI#Machine Learning#Performance Modeling#SoC#LLM#Cloud Data Server