CSCosmos
A searchable computer-science learning hub with interactive modules across web, systems, security, AI, and infrastructure.
Problem
Interactive computer-science explanations are often isolated across small demos, making it difficult for learners to discover related concepts or understand what is available.
System
A typed topic catalog drives domain routes, search, active and planned states, native visualizers, external microsite links, and statically generated topic pages.
Proof
- Runs as a public Next.js application with a public topic catalog.
- Uses shared typed data to drive navigation, status, search, and topic routes.
- Combines native visualizers with focused external microsites.
Architecture
- CatalogTyped domain and topic data defines the learning map.
- DiscoverSearch and domain views surface relevant topics.
- ResolveStatus routes users to a native module or focused microsite.
- LearnInteractive visual systems make abstract concepts observable.
Decisions
- Keep the topic registry in TypeScript so route and status changes remain reviewable.
- Distinguish live modules from planned content instead of presenting placeholders as finished.
Tradeoffs and limitations
- Native integration improves continuity but increases the main repository surface.
- External microsites keep experiments isolated but distribute maintenance across deployments.
Technologies
- Next.js
- TypeScript
- React
- Static generation
- Vercel