Skip to content
View mingzheTerapines's full-sized avatar

Block or report mingzheTerapines

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
mingzheTerapines/README.md

Mingzhe Terapines

👋 Hi, I'm @mingzheTerapines, a software engineer with expertise in error analysis, infrastructure design, and C++ programming. Currently, I'm focused on MLIR (Multi-Level Intermediate Representation) to enhance my skills.

🎓 Education:

💼 Work Experience:

  • 2021-2022: Xiaomi Mobile Software Department - Camera Team(Camera Engineer major in Infra and Customize)
  • 2022-2024: Confidential Unit, Wuhan, China (Marine Engineer major in System Software)
  • 2024-Present: Terapine Technology, Wuhan, China (Software Engineer major in MLIR projects)

🔬 Projects:

  • CIRCT: An MLIR-based project for System Verilog simulation and synthesis, improving upon Yosys. 7 commits(1429++/126--)
    • Struct/Union complex type construction
    • Support Mem2Reg/SROA/canonicalizer/Dedup Pass
    • Support Verilator for E203
  • MAGIC: Graph Algorithm Compiler Project 23 commits (4278 ++/807--) Key contributions:
    • Completed full - link support for backends such as Tenstorrent:
    • Built the basic framework.
    • Developed the Driver Tool (magic - compile).
    • Implemented the dynamic front - end language adaptation mechanism.
    • Designed the core algorithm of the Cost Model and its code.
    • Completed the development of multiple operator Lowering Passes.
  • IREE-XXXX: IREE with some RVCPU-NPU device E2E compiler 185 commits (115,577 ++/46,839--)
  1. Architecture construction:
  • Built the IREE plug - in framework system IREE - ESWIN and wrote the relevant code.
  • Supported the HAL layer extension architecture.
  • Refactored the ETensorType.
  1. Key technologies:
  • Ported the XXXIR dialect and optimized the basic components of the Pass to form the XXXIR optimization pass.
  • Developed the quantization operator Lowering scheme.
  • Constructed the XXXNN basic framework.
  1. Project management:
  • Organized multiple cross - team/intra - team coordination meetings.
  • Assisted in completing end - to - end verification of multiple models. Supported 10+ models

💻 Skills & Interests:

  • MLIR
  • Infrastructure design
  • Error analysis
  • High-level programming languages

💌 Contact: simbazhang96@foxmail.com

💡 Fun Fact: I have a passion for exploring the intersection of software engineering and artificial intelligence.

✨ Keep learning, keep growing! Generated by AI.

Popular repositories Loading

  1. modern_comipler_learning modern_comipler_learning Public

    现代编译原理学习

    C++ 1

  2. modern_compiler_code modern_compiler_code Public

    现代编译原理原书代码

    C

  3. mingzheTerapines mingzheTerapines Public

    Config files for my GitHub profile.

  4. aliyunpanshare aliyunpanshare Public

    Forked from acoooder/aliyunpanshare

    阿里云盘、夸克网盘影视资源分享,每日发布最新电视剧、综艺、电影资源。

  5. mojo-learning mojo-learning Public

    Forked from rd4com/mojo-learning

    📖 Learn some mojo !