How team-duiduidui/geniusrank works
Project type and purpose
• This is a full-stack, microservice-based web application called GeniusRank.
• Purpose: provide GitHub user data viewing, comprehensive ability evaluation (ranking algorithms), nationality guessing via AI, and exportable score cards.
File structure
• Top-level folders:
• frontend – Remix.js React/TS web app + SSR/SSG on Cloudflare Pages
• backend – Maven multi-module Spring Cloud microservices
– gateway-service (API gateway)
– user-service (user registration, login, profile, file upload)
– analyze-service (ranking and nationality guessing, AI integration, MQ consumers/producers)
– aggregation-service (data aggregation)
– framework (common Spring Boot auto-configuration, database, web, error handling)
• Shared resources: /backend/resources database DDL and API definitions.
• Config files: application*.yaml, shardingsphere configs, Docker/Maven wrappers.
• docs folder with detailed architecture and algorithm designs.
Extracted architecture from README
• Frontend: Remix.js + Tailwind + i18next + Zod + Mantine; SSR/ISSG/SSG; Cloudflare Pages.
• Backend: Spring Boot 3 + SpringCloudAlibaba + Nacos (service registry/discovery) + MyBatis-Plus + Sharding-Sphere + Redis (cache) + MySQL (persistence) + RocketMQ (messaging) + XXL-JOB (scheduled tasks) + Sentinel (circuit breaking/rate limiting) + SkyWalking (tracing/monitoring).
• Interactions:
• Frontend ↔ API Gateway (HTTP/REST)
• API Gateway ↔ Services (REST/RPC via Feign + Nacos)
• Services ↔ MySQL/Redis
• Services ↔ RocketMQ for async decoupling
• XXL-JOB triggers tasks in user-service or analyze-service.