Starting point
I started this project because most language-learning apps felt too generic for what I wanted. They were good at repetition, but not very good at adapting to how I actually like to learn. I wanted something that could feel more practical, more personal, and more responsive than a fixed set of exercises. At the end of the day, I built it for my own needs: I wanted a better way to learn Japanese before my trip.
My role
This one is entirely my own. I designed and built the product, made the technical choices, and used it as a way to explore both language learning and AI-assisted development in a more hands-on way than reading or experimenting in isolation ever would have allowed.
What I worked on
I built a language-learning application where an AI tutor generates personalized lessons, tracks progress, and adjusts difficulty over time. I also added gamification elements because motivation matters just as much as lesson quality when the goal is to build a real habit.
From a technical point of view, the project gave me room to go deep on SvelteKit, AI workflows, and agent orchestration patterns. More importantly, it forced me to keep asking whether the product was actually becoming more useful or just more technically interesting. I wanted the app to stay grounded in a real learning need, not drift into being a demo for its own stack.
What made it tricky
The difficult part was making the AI feel like a genuine improvement instead of a gimmick. It is easy to build something that looks clever in a screenshot. It is much harder to build something you want to come back to every day. That tension shaped a lot of the product decisions.
The result
The result is an application that I use myself on a regular basis, which is the best signal I can ask for at this stage. It also gave me direct experience with AI agent orchestration patterns in the context of a real product instead of a detached technical exercise. Those lessons around AI workflows and orchestration have carried into later projects too, whenever they have been relevant.
What I took from it
I like building things that sit at the edge of curiosity and usefulness. This project reminded me that personal projects can be serious work too, especially when they force you to test your ideas against your own daily habits.