Looking back at building JAMC, a Q&A platform, and the lessons learned along the way. Reflections on technical decisions, challenges faced, and what I would do differently.
Throughout my own learning experience, I’ve always felt that public classrooms have a common issue: every teacher has to deal with a large group of students, each with their own learning pace. This mismatch makes it tough for everyone to learn effectively. Teachers try to solve this by splitting students into groups, but I always felt there might be a better way. Personally, I believe fostering the habit of asking questions is crucial for deep learning, and this became the inspiration for JAMC—a platform designed to encourage asking questions and help teachers keep track of student progress.
For the development, I chose Next.js simply because it was popular and I was familiar with it, which allowed me to build quickly. Honestly, scalability wasn’t something I thought too deeply about at first, and I’m okay with that choice—it suited what I needed at the moment.
Initially, I tried designing the UI myself, and frankly, it was pretty bad. After some frustration, I discovered Vercel’s v0.dev, which helped me significantly improve the UI without needing deep design skills. Through this experience, I realized that building frontend logic isn’t the hardest part—making the whole page visually appealing and user-friendly is the real challenge. Normally, there’d be a design team behind this, but as an indie developer, v0.dev was a big help.
Now that most of the core features are done, I’m facing the next big step: deployment and getting real users. Suddenly, I see this massive gap between my GitHub project and what I’d call a “real product.” It feels like stepping into startup territory, which honestly makes me feel pretty nervous.
Edtech in Vietnam is still quite new—there aren’t really big, dominating players, nor has the education system seen major shifts yet. But there’s definitely growing interest and many people trying new ideas. On the flip side, Q&A platforms like Stack Overflow seem to be losing ground because of AI tools like ChatGPT. That made me wonder if maybe people don’t even need traditional Q&A platforms anymore because AI can answer everything. However, I think there’s still room for platforms that smartly combine AI with human interaction. JAMC originally aims to connect teachers and students better, not replace teachers with AI.
My next goal is to launch a small private beta with a real class to see if students and teachers find it useful. I’m curious to see if integrating some AI-generated suggestions with teacher moderation could create a balanced, effective learning environment. Right now, I’m still figuring out if this project will stay just a thesis project or if I’ll continue developing it further.
Writing this first blog has been interesting—my thoughts felt scattered at first, but putting them down helped clarify things. Hopefully, this experience-sharing resonates with someone else, or maybe my future self, reflecting on how far I’ve come.