Benjamin Zhou
52 posts

Benjamin Zhou
@Way_Wander
My little journal for learning AI from scratch. Neuroscience Student @nyuniversity
New York City Katılım Haziran 2025
337 Takip Edilen112 Takipçiler

@zarazhangrui I don't quite see the point of making extravagant designs of a slideshow, when the most important thing of it is to actually present your content comprehensively, which most people fail to do so
English

This is technically not a "game" yet; let me break it down of how I did this:
1. I use Midjourney (or other AI for image generation) to generate pictures of characters I like.
2. Then I put the picture into HunYuan 3D by Tencent, and I put my picture in and generate 3D models, and most importantly, you need to export the right 3D model file (in this case, it's .fbx).
3. Then I used Claude Code to ask him to make an interface or HTML that shows all these 3D models, and you add features like rotating the 3D model and so on. I'm really surprised that, as long as you structure your thoughts and put ALL of your imagination to Claude Code, he will help you out, really!
I hope this helps
English

@Way_Wander how did you code the game ? asking because i vibe code with stuff that i know. did you use unity or something else?
English

Inspired by Dilum, I vibe-coded a game UI using Midjourney for 2D design, Tencent HunYuan for 3D rendering, and Claude Code for coding it all together!
I'll try adding animations to these characters and more lighting effects soon!
Dilum Sanjaya@DilumSanjaya
Vibe coded a game character selection screen Everything here was made with AI tools Nano Banana: character design + UI Tencent Hunyuan3D: image to 3D Gemini Pro: UI More details ↓
English

@otxdave_ You should search "Tencent HunYuan 3D" and it directs you to the website and you can sign up an account with your email.
The generation time span is relatively long, sometimes it lags, so give it a second to process!
English

@Way_Wander Wow
I tried but couldn’t 💔
Because of the tencent hunyuan, stuff I don’t even know what that is
English

💡I was surprised that, to build this interface, I only needed to describe the problem and propose a solution for Gemini to implement.
The most important part is creativity: Gemini didn't come up with separating upward and downward scrolling into two distinct moves, which initially made the vision tracker overly sensitive to tiny movements and caused excessive vibrations and shakiness on the screen.
It was truly about thinking and imagining ways to alleviate issues and improve step by step.
English

I built a touch-free mouse for my browser that turns my hand gestures into scrolls and clicks, using
@GoogleAIStudio
Here are some insights upon building 👇
English

@Way_Wander Physics /hard sciences is so much more enjoyable when it can be visualized...
English

my chemistry professor said, in front of 725 students during the lecture, that the traditional atomic structure (Bohr model) is not how atoms actually are.
having no way to visualize this, i turned to Gemini and was absolutely amazed by the quality and infinite space for imagination.
it compares the Bohr model atom with the quantum model at different energy levels.
i'm blessed to be a student in this AI era. 🙏
English

1. Many participants reported entities saying, “We’ve been waiting for you.”
This may be associated with the "spotlight effect," a cognitive bias where people believe they are being noticed and watched by others more than they are.
DMT may activate an egocentric brain region, making us feel like protagonists in an ideal world.

English

@BrandonLuuMD So much focus on protein and carbs, yet micronutrients like vitamin C can be the metabolic rate limiter.
It’s a reminder that fat loss isn’t just about willpower or exercise!
English

@Way_Wander Regarding data - there's not much to suggest that slow caffeine metabolizers are at an elevated CVD risk from coffee consumption. In fact, they appear to be at a lower risk compared to decaf drinkers and non drinkers. See this MA: sciencedirect.com/science/articl…

English










