Opening page…
Opening page…
AI · Cloud · Education
An engineering background, with a focus on AI products and solutions.
I like turning needs into things we can test. When something is unfamiliar, I study good work and try to reproduce it. Once I understand it, I add my own judgment, make improvements, and keep refining through feedback.
A continuous curve of exploration — from engineering to AI, campus to products.
2020.09
Stepped into ECE — learning how systems are built from the ground up.
2021.04
Discipline on the track mirrored the same focus I bring to building.
2021.07
First attempts to turn learning needs into real services and products.
2022.07
Competition projects became a way to validate ideas under real pressure.
2023.05
Learned leadership through coordination, responsibility, and team culture.
2024.08
Went deeper into AI, robotics, and intelligent systems.
2024.10
Learned how AI changes the way software is written and shipped.
2025.03
From tools to agents — exploring systems that can plan and act.
2025.03
Brought product and AI practice into real company workflows.
2026.07
Joined Alibaba Cloud as a management trainee — learning how AI products ship at scale.
Future
Continuing to learn at the edge of AI, products, and meaningful building.
Things I have worked on — and what I learned along the way.

Jingbian Tutoring is a tutoring service platform that connects families of primary and secondary school students with qualified university student tutors. The project aims to reduce information asymmetry and help families discover suitable tutors more efficiently.

A conversational AI Agent designed around Lalamove’s urban freight-ordering scenario. It understands natural-language or voice requests, extracts order parameters, clarifies missing information, recommends suitable vehicle types, and generates a structured order draft for user confirmation.

Atoms, formerly MetaGPT X (MGX), is a multi-agent AI Coding platform developed by DeepWisdom. It transforms natural-language requirements into functional applications through the collaboration of AI product managers, engineers, architects, and data analysts. As a Full-Stack AI Product Manager Intern, I worked on Builder Runtime and AI Infrastructure capabilities—including Secrets, Auth, AI Capability, Cloud & AI Balance, Overview, Preview, Restore, and Publish—to simplify backend configuration, model integration, cloud-resource management, and application deployment for non-technical builders.

A tractor-mounted embedded system that enables farmers to capture field observations through voice interaction, automatically associate recordings with GPS and task information, and generate structured MP3 and JSON records for later analysis.
Notes on AI, products, founding, and life.

一句“帮我拿瓶水”的背后,藏着从电力、GPU 到 Agent 与具身智能的整条产业链。

很多人还在用“模型强不强”来理解这一轮 AI 竞争,但我越来越觉得,真正决定下一阶段格局的,可能已经不是模型本身了。

最近我仔细读了两份报告,一份是由 Epoch AI(Google DeepMind 委托)完成的 “AI 2030 — 量化未来” 报告,另一份是华为战略研究院发布的 《智能…

你可能已经在朋友圈、科技圈、公众号甚至招聘网站里见过“Agent”或“AI Agent”这个词。它似乎成了一个新的热词——但具体是什么?和我们现在一天到晚在用的 AI软件(比…
Campus, work, founding days and travel — simple visual memories.
从 fork 模板开始,通过个人网站学习 HTML 和 CSS。逐页重新设计,也留下了页面风格尚未统一的学习痕迹。