ApplyAI

AI-powered job application assistant that tailors your CV, drafts cover letters, and tracks every application in one place.

Features

• Paste a vacancy URL or job description — ApplyAI parses the posting
• Upload your master CV PDF — AI extracts text and tailors it per role
• Generate a role-specific cover letter via Google Gemini
• Track application history with status labels (Applied, Interview, Rejected, No response)
• Filter and search past applications
• Duplicate detection warns if you already applied to the same role
• Copy or download tailored CV as PDF

Stack

Layer	Tech
Frontend	React 19, Vite, Tailwind v4 — apps/web/README.md
Backend	Hono on Bun — apps/api/README.md
Database	PostgreSQL + Drizzle ORM (packages/db)
AI	Google Gemini (API service in apps/api)
Tooling	Biome (lint/format), Knip, Bun test
Infra	Docker Compose

Monorepo structure

applyAI/
├── apps/
│   ├── api/          # Hono REST API — see [apps/api/README.md](apps/api/README.md)
│   └── web/          # React SPA — see [apps/web/README.md](apps/web/README.md)
├── packages/
│   ├── db/           # Drizzle schema & migrations
│   └── shared/       # Zod schemas & shared types
└── docker-compose.yml

Quick start

Run with Docker (recommended)

Only Docker Desktop is required — no Bun install needed.

git clone https://github.com/irookanji/applyai.git
cd applyai
./start.sh

On first run, start.sh creates .env from .env.example. Add your GEMINI_API_KEY from Google AI Studio, then run ./start.sh again.

Open http://localhost:5173 when the containers are up. Stop with Ctrl+C.

Equivalent without the helper script:

cp .env.example .env   # add GEMINI_API_KEY
docker compose up --build

Local development

For day-to-day coding, use Bun and run Postgres in Docker:

Prerequisites: Bun, Docker Desktop

cp .env.example .env   # add GEMINI_API_KEY
bun run db:up
bun install
bun run db:migrate

# One terminal — API + web (stops Docker api/web if still running from ./start.sh)
bun run dev

# Or split across two terminals:
# bun run dev:api
# bun run dev:web

Open http://localhost:5173 (web docs). Vite proxies /api to port 3000 (API docs).

If you previously ran ./start.sh, stop it with Ctrl+C or run bun run docker:stop-app before local dev. bun run dev does this automatically.

Testing

Tests use Bun's built-in Jest-compatible runner (docs). Co-located *.test.ts files live next to the code they cover.

bun run test           # All tests in apps/ and packages/
bun run test:watch     # Watch mode
bun run test:api       # API only — see [apps/api/README.md](apps/api/README.md)
bun run test:shared    # Shared package tests only

Per-workspace docs: apps/web/README.md (UX flow) · apps/api/README.md (tests) · packages/shared (Zod schemas, date/hash helpers)

CI

GitHub Actions runs on every push and pull request to main (see .github/workflows/ci.yml):

1. bun install --frozen-lockfile
2. bun run lint — Biome
3. bun test apps packages — Bun test runner
4. bun run build — TypeScript + Vite production build
5. bun run knip — unused code and dependencies

Run the same pipeline locally: bun run ci

Portfolio notes

This project demonstrates:

• End-to-end AI integration with structured LLM outputs
• Monorepo architecture with shared validation schemas
• Preact Signals for UI mode state + TanStack Query for server state
• PDF ingestion and client-side PDF export
• Bun-native API tests with co-located unit tests
• Practical UX patterns (history-first, no dead-end screens, duplicate warnings)