Skip to content
View congkaili0123-slience's full-sized avatar

Block or report congkaili0123-slience

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.

Content in all repositories owned by your account will be closed.
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

Kasper Li · AI Product Builder

Builds AI enabled tools for real business workflows.

Turns practical problems into runnable products: maps users and constraints, defines product logic, and applies code and AI to improve the workflow.

Seeking 2027 graduate roles in AI Product Management, AI Applications and AI Solutions in Shenzhen or Hong Kong.

中文版本 / Chinese version

Selected Projects

An early stage B2B product for admissions planning organisations. It turns fragmented policy information and student context into explainable planning suggestions that consultants can review and communicate.

Defined the target users, B2B first strategy, product scope, matching variables and feasible AI workflow. Wrote the PRD and project plan, led market research, brand positioning, visual direction, UI architecture and team allocation. Coordinated the team to deliver a runnable end to end demo, now under refinement for future commercial validation.

Focus: product definition · business planning · workflow design · AI boundaries · team coordination

An independent scheduling and lesson management tool built from my experience as a private teacher. It lets parents choose from teacher defined availability and handles fixed lessons, rescheduling and lesson status through explicit rules.

Redesigned repeated scheduling coordination into a self service workflow. Designed, built and iterated the product through direct use in teaching practice.

Stack: Node.js · Express · multi tenant workflow · API tests · Playwright UI checks

Learning Analysis and Lesson Feedback System

An AI assisted workflow that turns a teacher's observations into structured learning analysis and post lesson feedback. It is based on a recurring service problem: parents need to understand what happened in a lesson, students need a record for review, and teachers need a reliable basis for the next lesson.

Structured the workflow around lesson observations, learning analysis and post lesson feedback. Applied retrieval augmented generation, prompt design and DeepSeek, then iterated the tool through direct use in teaching practice. This remains a personal working tool without external user or revenue data.

A browser based tool that maps an experience library to a job description and produces a tailored draft. Built for personal job search, with local only storage and no backend account system.

What I Bring

  • Business understanding: five years of direct K12 teaching experience and close contact with students, parents and service delivery.
  • Product judgment: Defines scope, separates rules from model generated content, and makes uncertainty visible when the cost of a wrong answer is high.
  • Hands on delivery: Moves from a vague workflow problem to a product decision, interface and working prototype.
  • Cross functional collaboration: Owns product decisions while keeping technical and delivery boundaries clear.

Background

  • City University of Hong Kong · MSc in AI-Driven Innovation · expected 2027
  • Shenzhen MSU-BIT · BEng in Electronic and Computer Engineering · GPA 3.72 / 4.0
    • Academic honors: Outstanding Student; multiple Academic Scholarships
  • Co-author of a paper published at IEEE CloudCom 2025
  • 5,000+ hours of K12 teaching experience

Contact

Email: congkaili0123@outlook.com


中文版本

关注 AI 如何进入真实业务流程,解决实际问题,并提升整体交付能力。

从具体问题出发:识别反复协调、信息难以复用和服务难以稳定交付的环节,明确用户、约束和产品逻辑,再结合代码与 AI 形成可以运行和验证的工具。

目前正在寻找深圳或香港的 2027 届 AI 产品经理、AI 应用产品和 AI 解决方案方向岗位。

代表项目

面向升学规划机构的早期 B 端产品。它把分散的政策信息和学生情况整理到同一条工作流中,输出便于顾问复核和沟通的升学规划建议。

担任产品负责人,定义目标用户、B 端优先路线、产品范围、匹配变量和 AI 工作流;完成 PRD、项目计划书、市场分析、品牌定位、视觉方向、UI 架构和人员分工。协调团队共同完成并跑通端到端 Demo,目前正在为后续商业验证进行优化。

工作重点: 产品定义 · 商业规划 · 流程设计 · AI 边界 · 团队协作

基于独立教师实际工作开发的排课与课时管理工具。老师开放可排时段,家长可以自主选课;固定上课、调课和课时状态通过明确规则管理。

将反复确认时间的沟通过程改造成自助式流程。独立完成产品设计与代码实现,并在教学实践中持续使用和迭代。

技术: Node.js · Express · 多租户流程 · 接口测试 · Playwright UI 检查

学情分析与课后反馈系统

将教师对学生课堂、作业和学习状态的观察,整理为结构化的学情分析与课后反馈。它来自一个持续存在的服务问题:家长需要了解课堂发生了什么,学生需要复习记录,教师需要依据反馈准备下一节课。

围绕课堂观察、学情分析和课后反馈定义完整流程,在教学实践中持续使用和迭代,实际采用 RAG、Prompt 设计和 DeepSeek。这是个人工作工具,目前没有外部用户或收入数据。

将个人经历素材与岗位要求逐条对应,并生成定制化简历草稿的浏览器工具。服务于个人求职过程,采用本地存储,没有后端账号系统。

我能带来的能力

  • 理解业务: 扎根五年一线 K12 教学,持续接触学生、家长和真实服务交付。
  • 产品判断: 明确范围,区分规则判断与模型生成内容,在错误代价较高时呈现不确定性。
  • 动手交付: 将模糊的流程问题推进到产品决策、界面和可运行原型。
  • 跨职能协作: 负责产品判断,尊重技术和交付边界,推动团队共同完成目标。

个人背景

  • 香港城市大学 · 人工智能驱动创新硕士 · 预计 2027 年毕业
  • 深圳北理莫斯科大学 · 电子与计算机工程本科 · GPA 3.72 / 4.0
    • 荣誉:优秀学生;多次获学业奖学金
  • IEEE CloudCom 2025 论文共同作者
  • K12 一线教学累计 5,000+ 小时

联系方式

邮箱:congkaili0123@outlook.com

Popular repositories Loading

  1. class-scheduler class-scheduler Public

    独立老师一对一排课系统:老师开放时段,家长自助选课,课时与调课规则自动结算。Node/Express + 原生 JS,含 44 项接口测试与真实浏览器 UI 检查。

    JavaScript

  2. admission-policy-matcher admission-policy-matcher Public

    基于用户画像的升学政策智能匹配系统(广东高考·B 端机构工具):产品定义、UI 架构与端到端 Demo。团队项目,本仓库只收录本人负责的产品设计与演示部分。

  3. resume-jd-matcher resume-jd-matcher Public

    简历适配工作台:把简历按目标 JD 逐条对齐的单文件工具。纯前端零依赖,打开即用。

    HTML

  4. congkaili0123-slience congkaili0123-slience Public

    个人主页