How udit-mimani/lendermatchplatform works
Project Type and Purpose
• This is a full-stack web application called LenderMatch AI.
• Purpose: match equipment finance loan applications to eligible lenders and programs, calculate fit scores, explain rejections, and persist/rerun applications.
File Structure Analysis
• Top-level folders:
• LenderMatch.API/ – .NET 8 Web API (Controllers, Services, Data, Entities, Migrations, config, Dockerfile)
• client/ – React 19 + TypeScript + Vite frontend (src, public, Dockerfile, nginx.conf, lint & TS configs)
• docker-compose.yml – orchestrates api, client (with Nginx proxy), and PostgreSQL 15
• LenderMatch.slnx – Visual Studio solution
• Architectural patterns: layered backend (Controller → Service → DbContext/Entities), SPA front-end served via Nginx, containerized with Docker Compose.
Key Insights from README
• CLIENT (React+TS) on port 3000 behind Nginx
• API (ASP.NET Core Web API + EF Core) on port 8080
• PostgreSQL database on port 5432
• REST endpoints in MatchController invoke MatchingService which uses EF Core to query/persist data.
a. Main Components
• User’s Browser / React SPA (port 3000)
• Nginx reverse-proxy + static server
• ASP.NET Core Web API (port 8080)
• MatchingService (business logic layer)
• EF Core / AppDbContext (data access layer)
• PostgreSQL database (port 5432)
• Docker Compose (orchestration)
b. Relationships & Data Flow
1. User interacts with React UI →
2. React issues HTTP/REST calls to Nginx →
3. Nginx forwards /api/* to .NET API (other paths serve static assets) →
4. MatchController validates input, calls MatchingService →
5. MatchingService derives features, applies lender/program rules →
6. EF Core persists/retrieves data via Npgsql to PostgreSQL →
7. API returns JSON results to React →
8. React renders match results or dashboards.
c. Architectural Patterns & Principles
• Layered architecture (Presentation → API/Controllers → Services → Data Access)
• Single Page Application (SPA) front-end separated from backend
• Containerized deployment with Docker Compose (loose coupling, easy scaling)
• Configuration via environment variables and appsettings.json
• Infrastructure as code for local vs. production (docker-compose.yml, launchSettings.json)
d. Relevant Technologies
• Frontend: React 19, TypeScript, Vite, Tailwind CSS, Axios
• Backend: .NET 8, ASP.NET Core Web API, EF Core 8, Npgsql
• Database: PostgreSQL 15
• DevOps: Docker, Docker Compose, Nginx