Now
24 january 2025,
We’ve welcomed a new member to our home in November 2024: Mangostine (Măng Cụt) - a ridgeback Phu Quoc dog. He’s growing so fast.
Learning chinese
I really like the way the Chinese language works. Each week, I try to learn a few words with Hanzi.live, a web app that I’ve built for my language learning process.
It is intended to be used to access the knowledge tree of the language. I’ve condensed the grammar of the language into various types of sentences and words. It’s in Vietnamese though. It will stay free and ad-free as long as it exists. Probably, I will open-source it someday.
I’ve translated the Tao Te Ching in both Vietnamese and English. I’m quite happy with it.
Making music
Since March 2024, I’ve had to temporarily halt the process of composing a piano concerto and a symphony. Instead, with Jane and Dawn, we’ve decided to make a RPG video game in 2D, and mostly because it’s fun, and doing things together is even more fun. And, a game needs music too after all.
Making a video game
Details on the game here.
Starting February 2025, after the lunar new year, things will move pretty fast with the help of our LLM and LDM AI-assistants, once they’re fully set up for our workflows.
Playing with Local AI
I’m experimenting with local llms (large language models). Some impressive local LLMs: Qwen2.5, Gemma2, Phi4, and recently Deepseek. I use Ollama to get things running and Open WebUI as the interface. It’s quite a fascinating time to be experimenting with local LLMs. I’ve set up a few tuned llms for my process of writing the story for the game. It’s quite cool to be able to run a large language model on your own machine, even without internet. Besides, I’m tweaking with the Latent Diffusion Model (Stable Diffusion, Flux) using the interface of ComfyUI, for image generation.
Things that I like:
Things that I use: