AP student · Fuzhou, China

I have a soft spot for turning wild ideas into things that actually exist.

I'm Venus Chen, a high-school student from Fuzhou.

I've built an AP Economics study product and an IELTS Speaking tool; served as president of my school's business club and dragged my classmates out to sell flowers off campus for the first time; and rehearsed and performed with my own band.

I can't really write code in the traditional sense, but I love vibe coding — not because I want to become a programmer, but because I can't stand watching a good idea die in my head.

So I keep asking: if this thing actually got built, would anyone want to use it?

Sometimes the answer is "yes." Sometimes it's "absolutely not." I kind of love both answers.

商模社社长把商业从 PPT 搬到街头 乐队发起人排练、演出,然后继续写歌 4 个产品上线从“我有个想法”到“真的有人能打开”

02 — Business Club · 纳思商业模拟社

Learning business by actually doing it.

I'm not a fan of talking business only inside a classroom. Over the past thirteen months we've run three stalls: two Qixi flower sales and a tarot booth at the sports-meet flea market. Every time we started with sourcing and pricing, and ended with costing, a written retrospective, and profit-sharing under an agreement written up front. I want the club's "business" to live somewhere beyond PowerPoint.

On campus: teach them to build, then let them build

I run a year-long elective under the club's name — 40 minutes a week, taught by me, with 50+ students enrolled. I didn't want it to be "listen to me lecture about business." I want every team to live through the birth of a product firsthand: find the problem → find the users → build the product → figure out the business model → bring in AI → ship it → get real people to test it → keep iterating. Each group has three roles — Explorer, Builder, Product Owner: someone hunts problems and users, someone builds the thing, someone keeps the whole product moving — and it all ends with a Demo Day where they actually show what they made.

The funny thing is, the course design isn't what made me happiest. Before class started I asked everyone: "what do you most want to learn?" The notes came back: economics, AI coding, monopoly simulations — and one said "I want to learn all of it." Since then the classroom has been full almost every week. It was the first time I realized that teaching people to build is itself a product: you have to hear what they actually want before deciding what to teach.

Off campus: business can't stay in the classroom

From Qixi 2025 to the 2026 sports meet, the club ran three stalls in total: two flower sales and one tarot-reading booth. Before each stall we worked out costs, sourcing, pricing and a profit target; afterwards we reconciled every income and expense line, wrote the retrospective, and split profits under a written agreement.

At the first flower sale, what I cared about most was "can we sell it all?" By the later stalls, I cared about something else more: can a teenager actually learn how business works from running a street stall?

Inside the elective

Teaching the vibe coding elective at the whiteboard A full classroom of students in the elective Student wish-list notes: AI coding, economics, monopoly — and one wrote I want to learn all of it

Stall #1: Qixi 2025 flower sale

Aug 29, 2025 · first off-campus stall
¥415.9cost
¥98profit

Fresh flowers, handmade bracelets, and a polaroid corner. The margin was modest, but we ran the full loop — sourcing, pricing, selling, and splitting the take under a written agreement.

Club members at the night flower stall, Qixi 2025 Handwritten price board with bracelets and flowers, Qixi 2025 The stall chalkboard at night, Qixi 2025

Stall #2: Qixi 2026 flower sale

Aug 19, 2026 · 7 hours near Dongbai Center, Fuzhou
¥382.19cost
≈¥697revenue · 55 transactions
45.17%profit margin (target 30%)
100+stems, nearly sold out

Seven people split into departments — procurement, promotion, sales, finance — under a written profit-sharing plan. Compared with our first stall a year earlier, the scale, the division of labor, and the mechanics were all a level more complete.

What actually sold: single stems and flowers customers assembled on the spot. What didn't: pre-designed bouquets, our mini-games, and the polaroid add-on. The plan survived contact with the street for about an hour.

The seven-member crew after the Qixi 2026 sale Wheeling the flower cart through the night market, Qixi 2026 Flower cart with the 6.6-yuan price boards, Qixi 2026

Stall #3: tarot readings at the sports-meet flea market

Sep 30, 2026 · school sports-meet flea market
¥13.7cost
¥163.32total profit

For the third stall we changed the product: tarot readings at the flea market. Nearly zero cost, and pricing works completely differently when what you sell is an experience, not a bouquet.

Tarot booth poster and deck at the flea market The tarot reading mat in action at the flea market

"Business isn't one sale that works. It's turning every miss into next time's playbook."

03 — Music

极光乐队(Aurora Band)

我在学校组建的乐队。2025 年 9 月第一次登台,九个月里完成了三场校内演出——从新生文艺汇演,到跨年元旦晚会,再到夏季音乐会。

乐队档案

  • 队名极光乐队 · Aurora Band
  • 成立2025 · 福州八中
  • 成员Lucian 主唱/键盘 · Kobe 贝斯 · Jennie 鼓 · Jermy 木吉他 · Henry 电吉他 · Venus 钢琴/木吉他/葫芦丝
  • 我的角色发起人 · 钢琴、木吉他、葫芦丝

三场校内演出

  1. 2025.9.20 · 文艺汇演(高一新生文艺汇演)
  2. 2025.12.31 · 元旦晚会
  3. 2026.6.11 · 夏季音乐会
2025.9.20文艺汇演(高一新生文艺汇演)
Aurora Band group photo after the freshman arts showcase On stage performing at the arts showcase
2025.12.31元旦晚会
Performing at the New Year's Eve gala Rehearsing in the studio
2026.6.11夏季音乐会
To the Summer concert program poster On stage at the summer concert

04 — Vibe Coding Products

Software I've shipped with AI as my coding partner.

"Vibe coding" is how I build: I describe the need, shape the product, and let AI handle the typing. Four products are online; one of them now has real users beyond my friends.

KiboTalk

Most Sincere Award AdventureX 2026

Practice for high-pressure real conversations — interviews, service counters — not just vocabulary drills.

I was the product owner on a four-person team: product direction, user scenarios, pitch narrative, and visual assets. For online interviews you use the app; for in-person ones, it optionally pairs with AR glasses for live translation and answer cues — the glasses are optional, never required. The event drew ≈8,975 applicants and 800 on-site; we won the Ant Group Lingguang track "Most Sincere Award" (¥2,000 team prize).

Built at the hackathon — demo on request.
The Ant Group Lingguang track Most Sincere Award for KiboTalk The KiboTalk app running a store-clerk conversation practice Pitching KiboTalk to visitors at the AdventureX booth The four-person KiboTalk team at AdventureX

PromptSpeak

Paused 32 users

IELTS speaking mock exams with scoring.

My first product, started September 2025. It worked — about 32 people used it — but acquisition cost was too high and competitors were entrenched, so I paused it. The pause itself was the finding: I now know what "too expensive to grow" feels like from the inside.

PromptSpeak scoring a mock answer: Band 7.5 overall with per-dimension scores The PromptSpeak sign-in screen
promptspeak.space ↗

HungMan

Live First real user story

An English vocabulary hangman that someone actually used to study.

A junior from school asked for a web version to memorize words, played it through, told me it beat rote memorization — and filed a feature request: custom wordlists need Chinese glosses. One real user with one real request taught me more than any launch metric.

hangman-english.pages.dev ↗

Quanzhou Heritage Runner

Side project

A vertical pixel-art runner featuring Quanzhou's intangible cultural heritage.

Built as part of a small-tools pack for Xiaohongshu — my first try at packaging a game for a content platform rather than for an app store.