名字只是起点，意义还在生长。a name left open to meaning
观察技术，也观察人。保持好奇，把想法做出来。Watch the technology, watch the people. Stay curious, and ship the ideas.
我是 Lumi Ruvenne。喜欢把模糊的问题拆开，把奇怪的想法做成能被体验的东西。不急着给自己一个固定标签——稳定的是内核，不是身份。I'm Lumi Ruvenne. I like pulling fuzzy problems apart and turning odd ideas into things you can actually experience. No rush for a fixed label — what's stable is the core, not the identity.
能力不是评分，是被项目反复验证的路径。Capabilities are paths repeatedly tested through projects.
我在意工具能不能让人少绕一点路、少担心一点，同时把关键决定留给人。长辈手机助手、隐私优先的求职工具，都是我正在做的尝试。I care about whether a tool can help people take fewer wrong turns and worry a little less, while keeping important decisions with people. A senior-friendly phone assistant and a privacy-first job tool are two projects I'm working on.
健康、陪伴、学习、家庭。不是宏大的未来，而是每天都会遇到、也会一直影响人的小问题。Health, companionship, learning, and family. Not a grand future, but small problems people meet every day and keep carrying with them.
先观察，再做出来；用测试和失败场景检查，最后记录哪里需要重来。Observe first, build it, test it against failure cases, then record what needs another pass.
健康、陪伴、学习、家庭。不是宏大的未来，而是每天都会遇到、也会一直影响人的小问题。Health, companionship, learning, and family. Not a grand future, but small problems people meet every day and keep carrying with them.
先观察，再做出来；用测试和失败场景检查，最后记录哪里需要重来。Observe first, build it, test it against failure cases, then record what needs another pass.
把复杂任务拆成可编排、可观察、可回滚的工作流，而不是只做一个聊天框。Turning complex tasks into observable, recoverable workflows instead of another chat box.
从文档解析、分块、嵌入到混合检索，关注真实召回质量和工程约束。Building retrieval pipelines around parsing, chunking, embeddings, hybrid search, and usable recall.
每个项目先定义用户痛点和不做什么，再选择最小可验证的实现路径。Defining the problem, the non-goals, and the smallest useful product path before building.
用 AI 编程工具把产品决策转成可运行项目，并依据测试、CI、部署和演示结果决定是否验收。Using AI coding tools to turn product decisions into runnable work, then making acceptance calls from tests, CI, deployment, and demos.
这个分类暂时没有项目，请选择其他分类或“全部”。No projects in this category yet. Choose another category or All.
把六面棋局、立体贪吃蛇、生活技能教学等一堆想法删到只剩一个 5×5 的空间经济谜题，再把稀缺性和取舍做成可以亲手操作的规则。Cutting a pile of ideas — six-sided boards, 3D snake, life-skill lessons — down to one 5×5 spatial-economic puzzle, turning scarcity and tradeoffs into rules a player can operate.
阅读案例 ↗Read case ↗从中英文网页、打印样式到单页 PDF，记录一次简历更新怎样追到实际下载文件，以及主题切换时的文字对比度问题。Following a resume revision through bilingual HTML, print styles, the downloadable PDF, and a theme-change contrast regression.
阅读案例 ↗Read case ↗把一份独立 HTML 接进网站时，修复脚本转换、方向键和进度保存问题，并把浏览器模拟与真实 Godot 实操分开。Integrating an HTML learning lab exposed script conversion, keyboard, and storage failures, alongside a gap between simulation and editor practice.
阅读案例 ↗Read case ↗一次二维码页面接入暴露了掩模运算和版本信息的位置错误。我们用独立解码器与参考矩阵，补上同实现往返测试的盲区。Mask arithmetic and version-information placement escaped internal round trips, prompting independent decoders and reference-matrix checks.
阅读案例 ↗Read case ↗从旧版自标记迁移到独立假名卡与 FSRS，再加上笔顺示范。复盘旧记录、音频失败、开源许可和动画生命周期怎样分别处理。Migrating self-marked progress to independent kana cards and FSRS, then adding stroke playback while protecting history and handling audio failures.
阅读案例 ↗Read case ↗从串场 PPT、人像和配音制作，到资料变更与会后交付，记录真实任务怎样逐渐沉淀成可复用的方法，以及哪些检查仍要由人完成。How real assignments involving slides, portraits, narration, and delivery became reusable methods, with clear limits on what AI and file checks can confirm.
阅读案例 ↗Read case ↗生活与成长Life & Growth
一个经济学背景的非技术创作者，如何从真实生活问题出发，借助 AI 和 Vibe Coding 开始构建、验证与交付。How a builder with an economics background began using AI and Vibe Coding to build, test, and deliver ideas rooted in everyday problems.
阅读笔记 ↗Read note ↗最近更新Latest造物手记Building Notes
收藏了很多设计网站之后，我把一句「如果作品集不只是整齐排列的卡片呢？」做成了一个可交互的空间作品集实验：空间探索 + 网格浏览，以及过程中关于设计判断、移动端与 AI Coding 的几点新认识。After collecting design websites for months, I turned one question — what if a portfolio wasn't a neat grid of cards? — into an interactive spatial portfolio experiment: spatial exploration plus grid browsing, and a few new understandings about design judgment, mobile, and AI coding.
阅读笔记 ↗Read note ↗最近更新Latest产品思考Product Thinking
AFTERWORDS v0.4 设计手记：由一首喜欢的歌和「词不达意」这个词出发，聊聊为什么表达不需要完美，以及一座小邮局为什么要给没说清楚的话留一个位置。AFTERWORDS v0.4 design note: starting from a favorite song and the idiom 「词不达意」 — words falling short — on why expression does not need to be perfect, and why a little post office should keep a place for the things we never quite say.
阅读笔记 ↗Read note ↗我学的是经济学，后来一边做项目，一边学写代码。做项目时，我先看问题、用户和约束，再决定 AI 应该介入哪里；做出来以后，再用测试和失败场景检查它到底靠不靠谱。I studied economics, then learned to code while building projects. When I build, I start with the problem, the user, and the constraints before deciding where AI belongs. Once something works, I test it against failure cases to see whether it actually holds up.
先判断问题是否真实，再决定做什么。Validate whether the problem is real before deciding what to build.
能把产品思路推进到可运行、可部署、可测试。Move product ideas into runnable, deployable, testable prototypes.
记录关键踩坑和取舍，不把项目包装成神话。Record tradeoffs and lessons instead of packaging projects as myths.
Contact
可以来 GitHub 看源码、发 Issue，或者和我聊产品、AI 工作流、内容自动化和岗位机会。也可以直接查看我的简历。Browse the source, open an issue, or reach out to discuss AI products, workflows, content automation, or roles. You can also view my resume directly.
