Skip to content

Commit

Permalink
refactor(ai4se-github-ai-platform): 精简并更新GitHub AI平台内容
Browse files Browse the repository at this point in the history
更新了GitHub AI平台的描述,简化了语言并添加了新的案例研究,包括GitHub Copilot对开发者效率的影响。同时,增加了CodeQL静态代码分析和Dependabot自动化依赖更新的相关内容,以及AI辅助研发决策框架的详细信息。这些更新旨在提供更清晰、更具体的AI在软件开发中的应用实例。
  • Loading branch information
phodal committed Aug 8, 2024
1 parent 1f754e9 commit cc41af0
Showing 1 changed file with 20 additions and 7 deletions.
27 changes: 20 additions & 7 deletions src/ai4se-github-ai-platform.md
Original file line number Diff line number Diff line change
@@ -1,14 +1,22 @@
# AI 辅助软件工程:GitHub 的 AI 平台工程赋能开发者

## 示例:GitHub
GitHub 的案例简单来说:围绕代码的开发、协作、构建为核心,以开发者体验作为度量体系

> 围绕代码的开发、协作、构建为核心,以开发者体验作为度量体系
引自官方的定义 [GitHub AI 驱动的开发者平台](https://github.blog/2023-11-08-universe-2023-copilot-transforms-github-into-the-ai-powered-developer-platform/)

[GitHub AI 驱动的开发者平台](https://github.blog/2023-11-08-universe-2023-copilot-transforms-github-into-the-ai-powered-developer-platform/)
## 辅助编码:GitHub + GitHub Copilot

### 工具、平台驱动
顶级发现显示,GitHub Copilot 有助于开发人员更快地编写代码,完成高达 46% 的代码,并使开发人员在工作中感到更加满足。

### [AI 增强的安全左移](https://github.blog/2023-11-08-ai-powered-appsec/)
- [GitHub Copilot 数据](https://github.blog/news-insights/research/the-economic-impact-of-the-ai-powered-developer-lifecycle-and-lessons-from-github-copilot/)
表明:开发人员需要差不多 3 个月才能达到的代码接受率基线水平(30%)
- 经验少的开发者更能受益

85% 的开发者在使用 GitHub Copilot 和 GitHub Copilot Chat 编写代码时,对代码质量更有信心。

[Research: Quantifying GitHub Copilot’s impact on code quality](https://github.blog/news-insights/research/research-quantifying-github-copilots-impact-on-code-quality/)

## [AI 增强的安全左移](https://github.blog/2023-11-08-ai-powered-appsec/)

![Images](images/github-security-shift.png)

Expand All @@ -21,7 +29,11 @@
- [CodeQL team uses AI to power vulnerability detection in code](https://github.blog/2023-09-12-codeql-team-uses-ai-to-power-vulnerability-detection-in-code/)
- [Multi-repository variant analysis: a powerful new way to perform security research across GitHub](https://github.blog/2023-03-09-multi-repository-variant-analysis-a-powerful-new-way-to-perform-security-research-across-github/)

### AI 辅助研发决策框架
### CodeQL 静态代码分析

### Dependabot 自动化依赖更新

## AI 辅助研发决策框架

[A developer’s second brain: Reducing complexity through partnership with AI](https://github.blog/2024-01-17-a-developers-second-brain-reducing-complexity-through-partnership-with-ai/)

Expand All @@ -40,7 +52,8 @@
3. **制定决策**。提出应该针对这个任务采取的措施。
4. **实现任务**。执行按照定义的顺序进行的步骤。

### 度量框架 [Space](https://queue.acm.org/detail.cfm?id=3454124)
## 度量框架 [Space](https://queue.acm.org/detail.cfm?id=3454124)

| | 满意度和幸福感 | 效率 | 活动 | 沟通和协作 | 效率和流程 |
|----|--------------------------|----------------|------------------|-------------------|-------------|
| 个体 | - 开发者满意度 | - 代码检视速度 | - 检视速度完成数量 | - 代码检视评分(质量或思考深度) | - 代码检视时长 |
Expand Down

0 comments on commit cc41af0

Please sign in to comment.