This is my first post on DEV, so I wanted to start by sharing a little bit about what I’ve been working on and, more importantly, why I started it.
Besides software development, I’ve been interested in Japanese for a while. I wanted to learn the language more seriously, especially vocabulary, but I kept running into the same problem: I couldn’t really find a learning workflow that fit the way I wanted to study.
I wanted something where I could save words while studying, review them later, take quizzes, keep track of what I was actually learning, and continue studying even when I didn’t have an internet connection.
I tried different approaches and applications, but I personally couldn’t stick with them.
At some point I thought: if I’m already a developer, why don’t I just build the thing I want to use myself?
That was basically how TangoBox started.
The original idea was quite simple: build a Japanese learning tool that fits my own way of studying. But as I worked on it, it naturally became much more interesting from a technical perspective.
One of the areas I’m most interested in is AI in general. I enjoy exploring different areas of AI, including image, video, audio and text processing, as well as LLM technologies.
For TangoBox, I wanted to explore what an LLM could look like when it is specifically adapted to Japanese learning instead of being a general-purpose chatbot.
Another thing that was important to me was offline learning.
I didn’t want the application to depend completely on an internet connection. The learning data, progress and other important information are kept on the user’s device. The chat model can also be downloaded to the device once, allowing the learning experience to continue without an internet connection.
For users who want to use more capable models, there is also the option to connect their own Gemini API key. This way, the user can choose how they want to interact with AI instead of the application forcing everyone into the same model or subscription.
There are also some simpler things that I personally wanted to have from the beginning:
Vocabulary saving and learning
Quizzes and learning progress tracking
Statistics and learning history
A Japanese-focused LLM experience
Offline learning capabilities
Hiragana and Katakana learning for complete beginners
Quizzes specifically for the Japanese alphabets
I’m still experimenting with the project and there are definitely things I want to improve. For me, that’s actually one of the most interesting parts of building it. It started as a solution to a personal problem and gradually turned into a project where I can experiment with mobile development, AI, local data, offline experiences and language learning at the same time.
For now, TangoBox is available only on Android.
There are no advertisements and no subscriptions. If you’re interested in trying it out, you can find it on Google Play:
This is my first DEV post, and I’m hoping to share more about the development process, the technical decisions behind the project, and the things I learn along the way.
Top comments (0)