Project Summary
An overview of the Job-Easy platform, its core domains, and technology stack.
Job-Easy is a complete application management platform designed to help job seekers discover, track, and apply to roles efficiently. Originally built as a simple CRM, it has evolved into a highly scalable, decoupled architecture containing 23 independent feature slices and 59 database models. It features automated job board polling, community-sourced global directories, and a native Gmail OAuth integration for precise MIME-threaded outreach.
Core Domains
Discovery & Polling
Users can upload private CSVs of companies, and opt-in to share them with the community. The platform also features a robust automated discovery engine that polls ATS job boards to find live postings, utilizing a PostgreSQL pg_trgm GIN index for millisecond-latency fuzzy search.
Gmail API & Outreach
We abandoned generic SMTP for a direct Gmail OAuth integration. The app manages AES-256-GCM encrypted refresh tokens and automatically synchronizes email threads with the user's Gmail inbox via Google Cloud Pub/Sub webhooks.
Application Tracking & Resumes
Users track their specific applications and manage Cloudinary-hosted resumes. HMAC-signed share tokens allow for temporary, secure external resume views, while automated cron jobs clean up stale attachments.
Platform Administration
A comprehensive admin layer handles user approvals, un-deletable audit logging via SetNull snapshots, support tickets, and platform health monitoring (including a custom circuit-breaker system for rate limits).
Technology Stack
- 1Next.js 15 & React 19
App Router paradigm, strictly enforcing the boundary between client and server components using
server-onlyand ESLint zones. - 2Tailwind CSS 4
Using the latest engine for styling, with a highly customized dark/light mode design system built on CSS variables.
- 3Prisma & PostgreSQL 16
Managing 59 highly normalized models, utilizing raw SQL extensions for SKIP LOCKED queues and trigram searches.
- 4Node:Test
Lightning-fast unit testing with the native test runner, alongside isolated database-backed integration harnesses.
Outdated Legacy Information