About Skellar
I’ve always been fascinated by the idea of a gamified learning platform. For a long time, it was just a thought due to the technical challenges. But recent advancements in AI have broken down many of those barriers. With a break between jobs, I finally decided to make this project a reality.
The core concept of Skellar is simple: transform learning into a skill map.
Imagine this: you complete tasks, unlock skills, and see your progress laid out clearly. You know exactly what you’ve learned and what’s next. You can even connect these achievements directly to your resume or portfolio.
What to Learn and How
My initial idea for Skellar was a huge skill tree where you could learn “everything.” I quickly realized this was too ambitious. I learned a key lesson about product development: constraints are a source of power. Just like how Duolingo focuses on languages and Brilliant on STEM, Skellar needs to start with a specific field to get the gamification right.
But I also didn’t want to compromise my vision. So, I split the platform into two parts:
The Skill Map: This is your personal skill tree, showing your progress and path.
The Playground: This is where you actually learn by doing. When you unlock a skill on the map, you’re taken to an environment to “learn by doing, playing.”
The map and the playground are separate experiences, but they’re linked together, creating a clear and engaging journey.
A New Approach to Coding
The first version of Skellar will launch with a coding course. People often ask, “How is this different from Codecademy or LeetCode?”
I thought back to my own learning experience. The most effective way I learned wasn’t by doing one exercise after another. It was by: starting with a completed project, trying to add new features, and then testing them. This forces you to understand the existing code and build something new at the same time.
This led me to design a unique learning model for Skellar:
The system provides a realistic codebase.
The content is broken into missions.
You develop a feature, test it, and submit a pull request (PR).
The system automatically verifies your PR with unit tests.
This experience is designed to feel as real as possible. The main challenge is making sure the learning curve is smooth and the interface is clear enough to prevent learners from getting stuck and quitting.
The platform will also be open to third-party creators, whom we call Starscribers, who can upload their own courses. The goal is to build a community based on two-way knowledge sharing. In the future, we could expand to all kinds of knowledge, even linking with real-world classes.
The MVP and the Future
Before the official launch, I’ve created a minimum viable product (MVP). You can check it out and give me feedback, or join the Starscribers Discord community.
- Frontend: Next.js + Shadcn UI
- Backend: Django + DRF
- Database: PostgreSQL
- The graph layout is created with ReactFlow.
I chose these technologies because they are mature, widely supported, and work well with AI tools like GitHub Copilot — which will be doing a lot of the work!
For me, Skellar is an experiment. I’m not focused on building the next “unicorn” product. I’m simply trying to answer one question: can learning truly feel like a game? I’m putting my idea out there and seeing what happens.
I don’t know what the final answer will be, but I know the journey will be fun. If this idea resonates with you, or if you’re curious about how a learning platform is built from scratch, I invite you to join the Discord and witness the growth of this skill constellation.
Starscribers Discord Server: https://discord.gg/KURmknGA
Skellar Open Beta: https://skellar.rn-ws.com/
Top comments (0)