jobberman 0.1.0

Autonomous AI-powered job search assistant CLI — scans, scores, tailors, and applies.

Jobberman — Agentic Job Search Assistant

An autonomous, AI-powered CLI tool that scans job boards, scores relevance, tailors your resume, and applies — all on autopilot, powered by Gemini.

       _       _     _
      | |     | |   | |
      | | ___ | |__ | |__   ___ _ __ _ __ ___   __ _ _ __
  _   | |/ _ \ | '_ \| '_ \ / _ \ '__| '_ ` _ \ / _` | '_ \
 | |__| | (_) | |_) | |_) |  __/ |  | | | | | | (_| | | | |
  \____/ \___/|_.__/|_.__/ \___|_|  |_| |_| |_|\__,_|_| |_|

---

Features

Feature	Description
Autopilot Mode	Continuously scans job boards, scores, tailors, and auto-applies
Gemini-Powered Scoring	Rates candidate–job fit 0–100 before wasting API calls
Resume & Cover Letter Tailoring	AI-generates tailored content per job description
Portfolio Coaching	Suggests projects to strengthen your application
Mock Interview Prep	Generates role-specific questions with feedback
LinkedIn Auto-Apply	Best-effort Easy Apply via Playwright automation
Storage Management	Auto-prunes packages, rotates logs for production
Deduplication	Never reprocesses the same job listing

---

Quick Start

1. Install

# Clone and install globally
git clone https://github.com/joey727/Jobberman && cd Jobberman
pip install -e .

2. Set Your API Key

export GEMINI_API_KEY="your-gemini-api-key"

Get one free at Google AI Studio.

3. Run a Single Job Analysis

job-search full \
  -j job_description.txt \
  -r resume.txt \
  --name "Jordan Doe" \
  --email "[e-mail]" \
  --headline "Backend Engineer" \
  --years-experience 5 \
  --skills "Python, Django, PostgreSQL, AWS"

4. Start Autopilot

job-search autopilot \
  --query "python backend" \
  --sources remoteok,remotive \
  --interval 30 \
  --min-score 60 \
  -r resume.txt \
  --name "Jordan Doe" \
  --email "[e-mail]" \
  --headline "Backend Engineer" \
  --years-experience 5 \
  --skills "Python, Django, PostgreSQL"

This will:

1. Scan RemoteOK + Remotive every 30 minutes
2. Score each job against your profile (skip below 60/100)
3. Tailor your resume + cover letter for qualifying jobs
4. Save consolidated application packages to autopilot_output/
5. Log every action to application_log.jsonl

Stop with Ctrl+C — graceful shutdown guaranteed.

---

Autopilot Output Structure

autopilot_output/
├── packages/
│   ├── 20260320T184500_senior-python-engineer.json    ← Single consolidated file
│   └── 20260320T185200_backend-developer-aws.json
├── applied_jobs.json       ← Dedup tracker (never reapply)
├── application_log.jsonl   ← Structured action log
└── application_log.old.jsonl  ← Rotated archive

Each package JSON contains the job description, score, tailored resume, and cover letter in one file.

Storage Management Flags

Flag	Default	Description
--max-packages	50	Auto-delete oldest packages beyond this limit
--max-log-lines	5000	Rotate log file when exceeding this line count

---

All CLI Commands

Command	Description
job-search full	Single job: parse JD → tailor resume → portfolio → interview
job-search autopilot	Continuous bot: scan → score → tailor → apply → log
job-search run	Batch scan from a source + tailor per job
job-search apply	LinkedIn Easy Apply from generated packages
job-search apply-and-reach-out	Apply + resolve company + collect contacts + draft emails
job-search collect-contacts	Scrape LinkedIn employee profiles
job-search message-contacts	Draft and optionally send outreach emails
job-search export-auth	Export Playwright authenticated session
job-search --version	Print version

---

LinkedIn Automation (Optional)

For LinkedIn auto-apply and contact scraping:

# 1. Install Playwright
python3 -m playwright install

# 2. Export your LinkedIn session
job-search export-auth \
  --login-url "https://www.linkedin.com/login" \
  --output-path storage_state.json

# 3. Use with autopilot
job-search autopilot \
  --query "python" \
  --storage-state storage_state.json \
  -r resume.txt --name "Jordan Doe" ...

Safety: LinkedIn Easy Apply submissions are auto-confirmed in autopilot mode. Manual apply commands still prompt for confirmation.

---

Tests

All 32 tests pass with no API key required (Gemini calls are fully mocked):

pip install pytest
pytest tests/ -v

Test File	Coverage
test_models.py	All 11 Pydantic models
test_scoring.py	Relevance scorer
test_bot.py	Bot loop, dedup, logging, pruning
test_agents.py	All 4 agents

---

Project Structure

job_search_agent/
├── agents/          # Gemini-powered agents (parser, tailor, portfolio, interview)
├── automation/      # Playwright automation (LinkedIn apply, contacts, email)
├── sources/         # Job board scanners (RemoteOK, Remotive, WeWorkRemotely)
├── bot.py           # Autonomous daemon loop
├── scoring.py       # Gemini relevance scorer
├── models.py        # Pydantic data models
├── orchestrator.py  # Multi-agent pipeline coordinator
├── cli.py           # CLI entry point
└── terminal_ui.py   # Rich terminal rendering
tests/               # Pytest suite (32 tests)

---

Deployment

Install from GitHub (Any Machine)

pip install git+https://github.com/joey727/Jobberman.git

That's it — job-search is now available globally. No cloning needed.

Install with pipx (Isolated Environment)

pipx install git+https://github.com/joey727/Jobberman.git

With LinkedIn Automation Support

pip install "jobberman[linkedin] @ git+https://github.com/joey727/Jobberman.git"
python3 -m playwright install

Publish to PyPI (Maintainer)

pip install build twine
python3 -m build
twine upload dist/*

Once published, anyone can install with:

pip install jobberman

---

Limitations

• LinkedIn DOM selectors change frequently — you may need to tweak automation selectors.
• Employee scraping and email extraction are best-effort.
• Screenshots are captured for diagnosing automation failures.
• This is automation scaffolding — always review outputs before sending.

License

MIT