Skip to content
View Liyao9797's full-sized avatar
🦄
Start with passion
🦄
Start with passion

Block or report Liyao9797

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.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

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

Report abuse
Liyao9797/README.md

Liyao|AI Product Manager in Progress

我正在从游戏策划转向 AI 产品经理方向,重点关注 AI Agent、游戏客服、玩家反馈结构化、A/B 测试分析和产品原型验证。

目前已完成的项目偏向“用可运行 MVP 证明产品判断”:从场景拆解、需求定义、流程设计、数据结构、评测指标,到通过 Codex 协作完成原型落地。

Featured Projects

Game Customer Support Agent MVP

game-customer-support-agent

一个面向游戏客服场景的 AI Agent MVP,验证“玩家自然语言反馈 -> Agent 追问补全 -> 结构化工单 -> 运营处理视图”的最小闭环。

我的角色:

  • 定义玩家反馈、客服处理、运营查看三类核心场景
  • 设计 Agent 意图识别、字段补全、优先级判断和状态流转
  • 设计反馈结构、知识库边界和评测脚本
  • 使用 Codex 协作完成可运行静态 MVP

关键词:AI Agent / Game Support / Feedback Structuring / Ops Workflow / Evaluation

Local A/B Test Workbench

ab-test-workbench

一个面向产品经理的本地 A/B 测试分析工具,支持 Excel 上传、显著性分析、质量 Gate、图表生成和 PM 可读结论输出。

我的角色:

  • 定义产品经理上传实验数据后的分析路径
  • 设计指标判断、显著性结果和风险提示表达
  • 将统计结果转成更容易用于产品决策的结论
  • 使用 FastAPI、React 和样例数据完成本地可运行原型

关键词:A/B Testing / Product Analytics / Data Analysis / Experiment Decision

Product Focus

  • AI Agent 产品原型设计
  • 游戏玩家反馈与客服场景
  • A/B 测试与产品数据分析
  • 产品需求拆解、流程设计、MVP 验证
  • 使用 AI Coding 工具将产品想法快速落地为可演示原型

Tech & Tools

Product: PRD / User Flow / MVP Scope / Evaluation Design
AI & Data: Agent Workflow / Knowledge Base / Feedback Schema / A/B Testing
Build: JavaScript / React / Python / FastAPI / Codex

Pinned Loading

  1. ab-test-workbench ab-test-workbench Public

    A local A/B testing analysis workbench for Excel uploads, significance testing, quality gates, and chart generation.

    Python

  2. game-customer-support-agent game-customer-support-agent Public

    Static MVP for a game customer-support Agent: player WebView, feedback structuring, operations workbench, and smoke tests.

    JavaScript