Search by job, company or skills

B

AI chip system software architect/engineer-NPU

3-5 Years

This job is no longer accepting applications

  • Posted a month ago

Job Description

Responsibilities

1. Track the latest trends, products and technical architecture of the GPGPU/NPU industry, analyze the company's internal business system needs, combine the self-developed chip architecture, and output the system software and hardware collaborative design plan 2. Responsible for the development and delivery of the self-developed AI acceleration chip system software stack, including chip device side Firmware/Host side driver/system management diagnostic tools 3. Responsible for the software and hardware collaborative verification of the NPU/CIM storage and calculation direction of the self-developed AI acceleration chip, performance analysis and optimization, and operator use case development based on programmable interfaces 4. Responsible for the software stack transplantation, adaptation and integration of the NPU IP, including the Firmware/driver/Runtime/SDK of the NPU IP.

Qualifications

1. Bachelor degree or above, computer, architecture and other related majors 2. Familiar with Linux C/C++ development, with a solid programming foundation 3. Have experience in GPGPU/NPU chip project research and development, familiar with Nvidia/AMD or domestic manufacturer GPU computing products, familiar with CUDA/ROCm and other GPU software stacks 4. Be familiar with computer architecture, have an in-depth understanding of deep learning, be familiar with at least one GPGPU/NPU/CIM storage and computing architecture, and understand its microarchitecture, main instruction set, etc. 5. Have experience in GPGPU/NPU Firmware/Linux driver development, and have actual development and debugging experience in memory management, Kernel task scheduling, RAS, etc. 6. Be familiar with the software stack transplantation and integration of third-party NPU IP, and perform performance analysis and optimization in conjunction with SoC architecture. Bonus points: 1. Have experience in AI-related R&D, familiar with CUDA programming, familiar with AI frameworks, and have an in-depth understanding of one or more network model structures 2. Good at cross-team communication and collaboration, with experience in project or team management.

More Info

Job Type:
Function:
Employment Type:

About Company

ByteDance is a technology company operating a range of content platforms that inform, educate, entertain and inspire people across languages, cultures, and geographies.
Dedicated to building global platforms of creation and interaction, ByteDance now has a portfolio of applications available in over 150 markets and 75 languages. For example, TikTok, Helo, Vigo Video, Douyin, and Huoshan.
Dedicated to building global platforms of creation and interaction, ByteDance now has a portfolio of applications available in over 150 markets and 75 languages. For example, TikTok, Helo, Vigo Video, Douyin, and Huoshan.

Job ID: 105540981