Hi! My name is Caroline Cheng Huang!

I’m a curious learner, maker, and storyteller who finds joy in connecting ideas, people, and possibilities. Whether through science, design, or community work, I’m drawn to projects that blend creativity with social impact.

My academic interest lies in using mathematics to solve social problems and exploring how science and technology shape society.

Beyond my projects, I love singing, building intricate paper models, and playing badminton — hobbies that bring me tranquility, focus, and joy.

I am currently serving as the Regional Outreach Manager at WiSTEM NGO, an organization dedicated to empowering and supporting women in STEM fields across the globe. WiSTEM works to increase the visibility and representation of women in science, technology, engineering, and mathematics by fostering a global community of passionate individuals, offering resources, and creating opportunities for mentorship and networking.

As the Outreach Manager, my primary responsibility is to connect with ambassadors worldwide. I organize onboarding calls, provide continuous support, and ensure that our ambassadors are equipped with the tools they need to succeed in their roles. Through these efforts, I help strengthen our global network, creating an environment where collaboration and growth are at the forefront.

Since stepping into this role, I have had the privilege of engaging with ambassadors from over 10 countries, including representatives from Africa, Europe, Asia, North America, and South America. These conversations have been an incredible opportunity to share experiences, offer support, and expand the reach of WiSTEM’s mission worldwide.

A paper model is a three-dimensional object made by cutting, folding, and assembling paper or cardstock. These models can range from simple shapes to intricate designs, often representing real-world objects, such as buildings, vehicles, or even abstract concepts. Paper models are created by following a set of templates or blueprints that guide the folding and cutting process.

This form of crafting allows for an easy and cost-effective way to create detailed models without the need for advanced tools or materials. Paper models can serve educational purposes, be used for decoration, or even as part of creative projects, fostering skills in design, patience, and manual dexterity.

For me, paper modeling is a form of tranquility — a hobby that helps me find inner peace and strength while providing a space for relaxing and reflective thinking.

Product Display

As the school’s Birthday Ambassador, I designed personalized posters for each student’s birthday.
Each piece draws inspiration from their personality, interests, or a quote that reflects them, translating those traits into colors, typography, and visual rhythm.
For me, these posters are more than greetings — they’re small rituals of recognition, moments where each person feels seen through design.

Every year, for each classmate’s birthday, I collect heartfelt wishes from friends, capturing moments of joy, warmth, and friendship. These messages are more than words — they are memories we share and cherish together. Here’s a sample collection of these special birthday blessings. 🎉

In the Yearbook project, I was in charge of layout design and visual storytelling.

It was more than a design task — it was an act of curating memory.

Each grade, club, and event had its own visual rhythm, shaped through spacing, perspective, and color.

I photographed over 400 students and teachers myself, hoping to preserve the texture of our shared time — honest, fleeting, and kind.

To truly understand the world of food delivery riders, I decided to step into their shoes.

For three intense days, I became a crowdsourced delivery rider, immersing myself in the daily rhythm of an industry powered by apps and algorithms. This wasn't just an experiment; it was a personal journey to answer a pressing question: In a job that promises freedom and flexibility, why do so many riders find themselves running faster and harder, caught in a cycle of what feels like working against themselves?

I wanted to move beyond the statistics and headlines to feel the reality—the pressure of the countdown timer, the strategies for outsmarting the system, and the physical toll of a city in constant motion. This auto-ethnography is my firsthand account of that journey. It’s the story of what I learned, felt, and witnessed while navigating the streets, the rules, and the hidden pressures that define this modern form of work.

Join me as I explore the thin line between the pursuit of freedom and the reality of self-exploitation—from the inside.

On my first day delivering food, everything felt unfamiliar. The app's operating system, the restaurants, the routes. Before going online, I took a one-hour training session on the Meituan app, where I learned traffic rules, how to accept orders, and the basic delivery process. As a crowdsourced rider, unlike full-time dedicated riders, I didn't have to attend morning meetings or meet a daily minimum number of orders. But the trade-off was that the quality of orders I could get was somewhat lower.

As soon as I left my residential compound, I switched to online mode. According to the tutorial, I was supposed to browse available orders on the app's homepage, then choose one based on distance and price. Nervously, I accepted my first order from a nearby restaurant. I followed the GPS, only to realize halfway that I was heading toward the customer's location instead of the restaurant. I had to turn back to pick up the meal. Luckily, it was a scheduled order, meaning it just had to be delivered within a set time window, so the delay didn't count as late. The app showed me the route, estimated arrival time, red-light waiting times, and even a countdown timer. Since it was just one order, about 3 kilometers away, I didn't feel too rushed and didn't run any red lights.

When I reached the destination, I was asked to place the food in a smart locker. Not knowing how to operate it, I asked another rider, who explained that riders themselves have to pay to store the meal, about 0.6 yuan.

After completing the first order, I accepted another one headed to a mall. On the way, the system automatically assigned me two more orders. Since it was my first day and I wasn't yet familiar with handling multiple orders, I declined them. New riders can refuse up to ten orders per day, while regular riders are limited to four chances. To avoid wasting refusals, after each order, I would immediately go offline, so the system wouldn't assign me anything automatically.

When I finally reached the mall, I had trouble finding a parking space. Security guards stopped me from leaving my e-bike where I wanted, so I tried to lift it onto the sidewalk. But the bike was heavy, and when I gave it some throttle to push, it lurched forward and flipped into a flower bed. The front panel cracked, the handlebars were covered in dirt, and my hands were muddy, too. As I struggled to lift it, the bike suddenly rose—another delivery rider and a guard had come over to help me. The rider parked it properly for me, then handed me tissues—first dry ones, then wet ones—seeing my hands were covered in mud.

It was the lunch rush, the busiest time of the day, and this fellow rider must also have been on a tight schedule. Yet he paused, risking lateness, fines, and a downgrade in his rating, just to help me. If it had been an ordinary passerby, I might not have been as touched. But in an industry where every second matters, his willingness to stop showed a generosity that deeply moved me.

Later, I accepted another long-distance order and went back online to see if the system would assign me any along-the-way deliveries. Otherwise, one order could take me an entire hour. But since I had already refused too many earlier, the system probably marked me as a "bad rider." No favorable orders came my way—only far, inconvenient ones.

On the second day, I adjusted my maximum order capacity to three, realizing that the system often bundled orders with similar routes. This gave me a stronger sense of urgency. Experienced riders shared strategies such as logging in near large commercial centers before peak hours to secure "good orders."

The system soon assigned me three simultaneous orders in a mall. Unfamiliar with the layout, I wandered in confusion, unable to locate the stores as some brand names differed from those shown in the app. After circling for a long time, I finally collected all three meals but then forgot where I had parked my bike. By the time I found it, only fifteen minutes remained, while the navigation predicted a sixteen-minute ride.

To save time, I pushed the bike to its maximum speed of 25 km/h. At a major intersection, the app showed a green light with forty seconds remaining. I attempted to make it across but, with the old bike already damaged from the previous day's fall, I arrived just as the light turned red. A giant "150" appeared on the app, indicating a red light wait of 150 seconds. Meanwhile, all three orders had less than five minutes remaining. Anxiety overwhelmed me, and I disregarded both traffic rules and sociological theories I had intended to "observe." I ran the red light, speeding into the residential area, only to have all three orders time out anyway, earning me nothing.

Later, I rented a faster electric bike to avoid such problems in the future. I suddenly realized this too was a form of self-exploitation: the platform did not urge me to upgrade, but I instinctively did so to meet its demands.

Other challenges soon followed: waiting endlessly for delayed meals at restaurants, deciding whether to deliver partial orders first, and sprinting through rain-slicked streets. To avoid losses, I learned to adopt tactics from other riders: cutting through alleys to save time, calling customers to wait by the elevator, or prematurely clicking "delivered" before physically arriving. These strategies reduced lateness but also heightened risk.

By the third day, I raised my maximum order capacity to four. Familiarity with the routes reduced some anxiety, but managing four simultaneous orders required intense mental calculations of pickup and delivery sequences. The system offered estimated times, but in reality, I had to adjust constantly depending on each restaurant's speed.

One particularly memorable order involved carrying a heavy case of twelve bottled waters and additional drinks—over 8 kilograms in total—up and down stairs in an old residential building with no elevator. My arms were left sore and bruised from the weight, all for a payment of just seven yuan.

In another case, I had to reach the 28th floor of a high-rise. The elevator was crowded with riders, stopping almost every floor. Some riders phoned customers to wait at the elevator entrance, saving precious minutes, so I learned to imitate this strategy. Yet even with such efforts, I still encountered late deliveries, sometimes asking customers if I could click "delivered" early, only to be rejected by the system's GPS restrictions.