Skills
.Work. You
Rest

platform-dev

Official

Cross-backend platform API design & implementation

Software Engineering#development#platform#distributed#torch#dtensor#cross-backend#mindspore
Authormindspore-ai
Version1.0.0
Installs0

System Documentation

What problem does it solve?

Enables consistent cross-platform platform APIs development for HyperParallel, unifying PyTorch and MindSpore backend support, DTensor extensions, and shared platform utilities.

Core Features & Use Cases

  • Cross-platform API design and implementation across Torch and MindSpore backends
  • Support for platform-level features: FSDP, HSDP, Pipeline Parallelism, and Activation Checkpoint
  • DTensorBase extension and platform-agnostic operations
  • Workflow governance and testing guidance for platform-layer development

Quick Start

Add a new Platform API to the base class, implement corresponding backend logic for Torch and MindSpore, and validate parity with cross-backend tests.

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: platform-dev
Download link: https://github.com/mindspore-ai/hyper-parallel/archive/main.zip#platform-dev

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.