Dilum Sanjaya Profile picture
Software Engineer | Building scientific & intelligent systems people dream of. 💡Experiments https://t.co/02nVJK3h1c https://t.co/fzTHdVlL8T 🎬Demos https://t.co/cOGQInCQ9n
May 9 6 tweets 2 min read
Fun interactive science app ideas | Part 3

Played around with generating 3D biological structures and made an app to explore them interactively

UI Design
GPT Images 2

Code
Gemini 3.1 Pro

More demos ↓ Don't know enough about these to say how accurate the 3D models are, but from what I can tell, at least the main structures are there. Tried both Tripo and Hunyuan3D. Tripo seems to give better results overall.
Apr 13 5 tweets 3 min read
Playing with map interactions.

Generated this UI concept with Nano Banana and brought it to life using Gemini Pro

Thinking about starting a series that explores aesthetic variations in data visualization UIs
More details ↓ Here's the original image from Nano Banana. changed few elements based on Gemini's implementation. Image
Mar 17 9 tweets 3 min read
Started bringing a smart home idea I generated with Nano Banana 2 to life
Here's the first look

Code: Gemini 3.1 Pro
3D assets: Tripo (not affiliated)

More details ↓ I wanted to create a smart home app so I could experiment with AI automation. I generated a few designs with Nano Banana 2. Here are some other designs it produced. Image
Image
Image
Image
Jan 6 6 tweets 3 min read
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 ↓ I asked Nano Banana to create a character selection screen and gave it one of the characters to match. These are some of the first results I got. I did not spend much time refining the prompt. Image
Image
Image
Image
Nov 28, 2025 10 tweets 4 min read
Got a bunch of questions about how I made this effect in Python, so here's a quick breakdown and the code↓ As always, here is the code link if you don't want to read through the steps😉
You can run it locally or in a Colab notebook. It isn't resource heavy, so it works perfectly even on the free Colab tier.


I also generated images for each step using nano banana and it actually did a great job.github.com/dilums/opencv-…