I Wanted to Share an Idea with the World
Jellyfish Climb began as an experiment in vibe coding. Since I was going to try it, I wanted to create something personally meaningful, so I chose game development—something I had been interested in for a long time.
My goal was to make a simple game that people of different ages and backgrounds could pick up and enjoy.
I remembered playing Flappy Bird, a remarkably simple game controlled with taps that still made players want to try again and again. I wanted Jellyfish Climb to have the same kind of accessibility: very few controls, but enough depth to keep you coming back.
From the beginning, I also intended to release the game publicly. I wanted to see how people around the world would respond to my idea and what kind of reaction it might receive.
Why a Jellyfish?
When choosing the subject of the game, I wanted something calming—something that could help players relax. A jellyfish was the first idea that came to mind.
I imagined a jellyfish gently floating upward from the bottom of the sea. Turning that image into a game sounded interesting, and it became the foundation of Jellyfish Climb's world.
The background music and sounds of wind, water, and bubbles were also selected with an underwater atmosphere in mind. Rather than creating a busy soundscape, I wanted to evoke a quiet ocean where a jellyfish floats alone and gradually rises toward the surface.
The Pulse Mechanic Was Born Through Playtesting
The pulse mechanic that exists today was not part of the original game. At first, tapping simply moved the jellyfish upward.
After playing it myself, however, I felt that the controls needed more excitement. If the timing of each tap affected how quickly the jellyfish climbed, the player's actions could connect directly to a faster finish and a higher score.
The way a jellyfish expands and contracts naturally matched this idea. Through playtesting, this became the pulse mechanic: players time their input to the jellyfish's movement to climb more effectively.
Balancing Plankton and Speed
The original scoring system added plankton points and a remaining-time bonus separately.
After several playtests, I discovered that continuing to collect plankton was more rewarding than finishing quickly because the time bonus was too small. This reduced the importance of using the pulse mechanic well and climbing efficiently.
Increasing the time bonus too much would create the opposite problem: players would ignore the plankton and rush directly to the goal. That would remove the decision between taking a detour for plankton and preserving time.
To create that tradeoff, I changed the system so that the points earned from plankton are multiplied by a time rate based on the remaining time. Reaching the goal quickly without collecting plankton produces no meaningful score, while collecting plankton and finishing quickly can result in a much higher score.
If players could keep their points after running out of time, collecting plankton indefinitely would once again become the best strategy. For that reason, the score becomes zero when time runs out.
Calm Atmosphere and Competitive Tension
Jellyfish Climb is intended to combine a relaxing atmosphere with the excitement of a game.
The score and Best Score systems allow friends to compare their results. Even within the quiet underwater setting, there is a sense of tension and competition when trying to reach a higher score.
The alternating upward and downward currents were added to prevent the game from becoming too repetitive and to introduce another strategic element. Together with plankton, pulse timing, and the time rate, the currents create moment-to-moment decisions about what the player should prioritize.
Bringing the Same Experience to Smartphones
Jellyfish Climb is built with HTML, CSS, and TypeScript. As a web game, I wanted it to be easily accessible from smartphones as well as computers.
I considered placing left and right controls on the screen, but on smaller displays, those buttons could obscure the jellyfish and the underwater scenery. Pressing a direction button while also focusing on pulse timing was difficult.
Instead, the smartphone version uses the device's tilt to control horizontal movement. Without separate direction buttons, the player can concentrate on timing each pulse.
Testing on real devices also revealed new challenges. The jellyfish's relative size and the density of plankton can vary depending on the screen. The smartphone version still tends to make plankton easier to collect, which can lead to higher scores than on a computer. Balancing the game so that similar scores can be achieved on both platforms remains an ongoing challenge.
Dividing the Work Between AI and the Creator
In this project, I have entrusted most of the code implementation and management to AI. My role is to test the completed result and decide whether the gameplay, visuals, balance, and music match the intended experience.
AI acts as the coder, while I take responsibility for the game's direction and the quality of the final experience.
One of the strengths of vibe coding is that it allowed me to express what I wanted to create without first mastering every detail of programming syntax and implementation.
However, communicating an idea does not always produce exactly what I imagined. Explaining the desired result clearly—and deciding whether the implementation is correct—can be difficult.
Using AI does not remove the need for knowledge. Without understanding the subject, it is difficult to give appropriate instructions or judge the result. While configuring the game's security, I had to research unfamiliar terms one by one. The experience taught me that even when AI handles the implementation, I still need to continue learning.
What I Learned by Publishing the Game
The game was initially published through GitHub Pages, which was relatively straightforward. The more difficult part came later: acquiring a custom domain and configuring security through Cloudflare.
I had to understand unfamiliar terminology and determine which settings were required and in what order. Adding Analytics also involved some code-related challenges, but I was able to resolve them by consulting AI.
Seeing traffic on the first day of the public release was especially memorable. Not every visit necessarily came from a real player, but the data showed that some visitors had started the game.
I have not yet received direct feedback, which is disappointing, but simply knowing that someone had actually played something I made gave me a real sense of progress.
SEO will remain an ongoing challenge. I want to continue exploring how far the game can reach and how it might become known to people around the world.
Treating Audio Rights with Care
Because AI was involved in the development process, I felt it was especially important to treat third-party rights—including audio rights—with care.
I reviewed the usage terms for each audio source, included the required credits, and provided links leading back to the original creators or distribution pages.
If the game helps more people discover the audio and the people who created it, I believe that can be a positive outcome. As someone who has been allowed to use these works, I intend to continue treating them respectfully and following their terms.
Growing the Game Gradually
I want to preserve the game's original concept: it should not become excessively difficult, and it should remain approachable for a wide range of players.
The first feature I would like to add is more plankton varieties. Differences between them and a small amount of randomness could add excitement and competition, giving players more reasons to return and play for longer.
After that, I am considering a mode where players can continue climbing and see how high the jellyfish can go. I plan to work on it once I have a clear idea of how it should function as a complete game mode.
Visual options such as a pixel-art skin would come at a later stage, after the game has reached a wider audience.
Rather than changing everything at once, I want to develop the game gradually while observing how players respond. The project is not primarily driven by commercial goals. I want to continue at a comfortable pace, giving my ideas time to develop and allowing the game to grow naturally.
I already feel that creating Jellyfish Climb was worthwhile. If the number of visitors grows and the game begins reaching people in more countries, I will consider that one form of success.
I also hope to create a place where players can share feedback directly. Receiving positive responses there would make the experience of creating the game feel even more rewarding.