梁爽 supercooledith

Hi there 👋😜🤗

my name is Shuang Liang, my english name is Edith, you can also call me supercool.

The Foundation: Electronic Design & Hardware Engineering (2017 - 2021)

During my undergraduate years at Chongqing University, I immersed myself in electronic design competitions and TI-related hardware projects.

End-to-End Practice: Mastered the full lifecycle of hardware development, including sensor integration, data sheet analysis, circuit design (RC matching), PCB layout, and embedded programming.
Award-Winning: Recognized with multiple national and provincial awards in technology competitions.

萌芽探索：电子设计与硬件基石 (2017 - 2021)

在重庆大学本科期间，我深耕电子信息领域，频繁参与各类大学生电子设计竞赛。这段经历让我完成了从理论到实践的全栈蜕变：

Academic Deep Dive: FPGA & Hardware-Software Co-design (2021 - 2024)

At Beihang University, my master's research focused on "Hardware Acceleration of Transformer Algorithms on FPGA Platforms".

Technical Core: Specialized in Verilog, pipeline design, and memory access optimization.
Design Philosophy: Developed a profound understanding of the symbiotic relationship between software and hardware—designing optimized hardware to carry advanced algorithms while tailoring algorithms to hardware constraints.

学术深耕：FPGA 与 Transformer 软硬件协同 (2021 - 2024)

在北京航空航天大学攻读硕士期间，我主攻《基于 FPGA 平台的 Transformer 算法硬件加速研究》。

硬核技术：深入实践 Verilog 语言、流水线设计及内存访问优化。
设计哲学：深刻理解了算法与硬件的共生关系。算法需要强力的硬件承载，而硬件需根据算法特性进行定制化优化。通过软硬件协同设计，实现了模型训练、量化到硬件加速的全链路优化。

Career Evolution: AIDC Infrastructure Expert & Product Leader (2023 - Present)

Currently serving as an AIDC Infrastructure Specialist, focusing on full-stack architecture for large-scale GPU clusters.

Infrastructure Expertise: Deep expertise in NVIDIA GPU ecosystems (A/H/B series, DGX, SuperPOD). Expert in high-speed networking (InfiniBand/RoCE/NVLink) and storage strategies (Block/File/Object), with a proven track record of optimizing TCO for 100M+ RMB computing center projects.
Product Leadership: Led the "0-to-1" development and operation of multiple AI Cloud Platforms, driving tens of millions in revenue through meticulous project management and continuous iterative optimization.

职业跨越：AIDC 基础设施专家与产品领航 (2023 - 至今)

目前专注于 AIDC（人工智能计算中心）基础设施全栈架构，致力于为大模型提供极致算力支撑。

基础设施架构：精通 NVIDIA 全系列 GPU (A/H/B) 及其生态硬件 (DGX/SuperPOD) 。擅长高速网络 (IB/RoCE/NVLink) 与多态存储的复杂组网策略，主导过亿级算力中心项目的系统选型与拓扑优化。
从 0 到 1 的产品力：曾主导多款 AI 算力云平台的研发与运营，实现了从产品设计、冷启动到千万级营收的完整闭环，具备卓越的项目管理与跨职能协作能力。