Auto Job Applier 🚀

Automatically apply to internships with AI! Just add your resume and run one command.

What It Does

1. Scrapes job boards (Jobright, Intern List) every 5 minutes
2. Ranks jobs based on your resume (AI scoring)
3. Shows jobs on beautiful web dashboard with fit scores
4. Auto-applies using Gemini AI + browser automation
5. Saves proof - screenshots and action logs for every application

Quick Start (3 Steps!)

Step 1: Add Your Resume

Create my_resume.json in this folder:

{
  "name": "Your Name",
  "contacts": {
    "email": "[email protected]",
    "phone": "+1-234-567-8900",
    "location": "Your City, State"
  },
  "education": [
    {
      "school": "Your University",
      "degree": "BS Computer Science",
      "grad": "2026"
    }
  ],
  "skills": ["Python", "JavaScript", "SQL", "React"],
  "projects": [
    {
      "name": "Cool Project",
      "tech": ["Python", "FastAPI"],
      "bullets": ["Built something awesome", "Improved by 50%"]
    }
  ],
  "constraints": {
    "locations": ["US-remote", "San Francisco"],
    "visa": "US Citizen",
    "preferred_roles": ["Software Engineer Intern"]
  }
}

Step 2: Configure API Keys

Copy .env.example to .env and add your keys:

cp .env.example .env

Edit .env:

• Add your Gemini API Key (get at aistudio.google.com)
• Add your Gemini API Key (get from aistudio.google.com)
• (Optional) Configure storage (GCS or S3) for saving application proofs

Step 3: Start Everything!

./start.sh

That's it! The script will:

• Build Docker containers
• Start PostgreSQL + Redis
• Start API + Workers
• Load your resume
• Begin monitoring jobs

Access the Dashboard

Open your browser and go to:

http://localhost:8000

You'll see a beautiful dashboard with:

• 📊 Real-time statistics
• 🎯 Job listings with fit scores
• 🏷️ Matched keywords
• ✓/✗ One-click approve/reject buttons

Also available:

• API Docs: http://localhost:8000/docs
• Health Check: http://localhost:8000/healthz

How It Works

1. Every 5 minutes, the system checks job boards for new postings
2. AI Ranking: Each job is scored against your resume (0-100%)
3. Dashboard Review: High-scoring jobs (≥65%) appear on the dashboard
4. Your Decision: Click ✓ Yes, Apply or ✗ No, Skip
5. Auto-Apply: If approved, Gemini AI fills out the application
6. Proof Saved: Screenshots and logs saved to artifacts/{job_id}/

Usage

View Logs

docker compose logs -f

Stop Services

docker compose stop

Restart

docker compose restart

Full Reset

docker compose down -v
./start.sh

How It Works

Job Sources → Scraper → Database → AI Ranking → Slack Notification
                                                        ↓
                                                    You Click Yes
                                                        ↓
                                          Gemini AI + Playwright Apply
                                                        ↓
                                              Save Proof + Logs

Features

✅ Auto-polling - Checks for new jobs every 5 minutes
✅ AI scoring - Ranks jobs 0-100% fit with your resume
✅ Web dashboard - Beautiful UI to review and approve jobs
✅ Auto-yes mode - Skip prompts for high-fit jobs (≥85%)
✅ Smart form filling - Gemini AI fills out applications
✅ Full audit trail - Screenshots + action logs
✅ Safety limits - Daily caps, allowlist, CAPTCHA detection
✅ Cloud storage - Saves proof to GCS/S3

Configuration

Enable Auto-Apply

Skip the manual review for high-scoring jobs:

curl -X POST http://localhost:8000/settings/auto-yes \
  -H "Content-Type: application/json" \
  -d '{"auto_yes_enabled": true, "auto_yes_threshold": 0.85}'

Adjust Thresholds

Edit .env:

• FIT_THRESHOLD=0.65 - Minimum score to show on dashboard
• AUTO_YES_THRESHOLD=0.85 - Auto-apply above this score
• MAX_SUBMITS_PER_DAY=10 - Daily application limit

Add More Job Sources

Edit app/jobs/ to add new scrapers.

Safety

• Allowlist only - Only applies to approved ATS domains
• Daily limits - Max 10 applications per day (configurable)
• CAPTCHA detection - Stops if CAPTCHA found (no bypass)
• Kill switch - Set KILL_SWITCH=true to stop all applications
• Truth only - Never fabricates resume data

Troubleshooting

Docker not running?

sudo systemctl start docker

No jobs showing up?

1. Wait 5 minutes (polling interval)
2. Check logs: docker compose logs -f worker
3. Update HTML selectors in app/jobs/jobright_fetcher.py if needed

Dashboard not loading?

1. Check API is running: curl http://localhost:8000/healthz
2. View API logs: docker compose logs -f api
3. Restart: docker compose restart api

Application failed?

1. Check artifacts/{job-id}/ for screenshots
2. View logs: docker compose logs -f worker
3. Try dry-run first: see API docs at /docs

What Gets Created

• Containers: postgres, redis, api, worker, beat
• Volumes: Database storage (persists after restart)
• Artifacts: ./artifacts/ - All application screenshots + logs
• User Data: ./user_data_dirs/ - Browser profiles per company

Tech Stack

• Backend: FastAPI + Python 3.11
• Database: PostgreSQL 15
• Cache: Redis 7
• Workers: Celery
• Browser: Playwright + Chromium
• AI: Gemini 2.5 Flash
• Notifications: Slack SDK
• Storage: GCS or S3

API Endpoints

• GET /healthz - Health check
• POST /profile - Update resume
• POST /settings/auto-yes - Configure auto-apply
• POST /dry-run/apply - Test without submitting
• GET /jobs - List jobs
• GET /applications - View history

See full docs at http://localhost:8000/docs

Files

• my_resume.json - Your resume (you create this)
• .env - Your API keys (copy from .env.example)
• docker compose.yml - Service definitions
• Dockerfile - Container build
• start.sh - One-command setup
• app/ - Python application code

Support

Questions? Check the API docs or view logs:

# API logs
docker compose logs -f api

# Worker logs
docker compose logs -f worker

# All logs
docker compose logs -f

---

Built with ❤️ for automated job hunting

⚠️ Disclaimer: For personal use only. Respect platform ToS. Only use truthful resume data.