志望動機の書き方

This project contains Python scripts to automate job searching and analysis using AI-powered web automation.

このプロジェクトはPythonを使って、AIの力で志望動機を書けることになります。

🚀 Quick Start

1. Install Dependencies:

browser-useをインストールして、実行できるようになって下さい。

# example command for reference
pip install browser-use langchain-google-genai python-dotenv langchain-openai langchain-core==0.3.49 langchain-community==0.3.17 langchain==0.3.21

pnpm i

2. Set up your Google API key in .env:

   GOOGLE_API_KEY="your_api_key_here"
   

3. Run the script:

Maybe you want to look at Optional section first, but if you want to run the script directly, you can do it like this:

python main.py

4. Optional:

• create input/志望動機_instructions.md , can use 志望動機_instructions.example.md as reference
• create input/about-me.md
• create input/companies.json in the format below:

{
    "working": "CompanyA",
    "backlog": {
        "CompanyA": {
            "name": "CompanyA",
            "urls": [
                "https://companyA.co.jp/",
                "https://companyA.co.jp/about",
                "https://companyA.co.jp/services"
            ]
        },
        "CompanyB": {
            "name": "CompanyB",
            "url": "https://companyB.com/"
        },
        "CompanyC": {
            "name": "CompanyC",
            "urls": [
                "https://companyC.io/",
                "https://companyC.io/usecase/",
                "https://companyC.io/about/"
            ]
        }
    }
}

Happy Job Hunting! 🚀

---

Draft

quick_job_research.py (🌟 Start Here)

• Purpose: Simple, focused job search on LinkedIn
• What it does: Searches for specific role, extracts skills/requirements, provides keyword analysis
• Best for: Getting started, testing the system, quick market research

simple_job_search.py (Advanced)

• Purpose: Multi-platform job search (LinkedIn, Indeed, AngelList)
• What it does: Searches multiple sites, compares results, generates comprehensive analysis
• Best for: Thorough market research, comparing opportunities across platforms

job_search.py (Full Featured)

• Purpose: Complete job search automation system
• What it does: Advanced search, detailed analysis, result saving, trending insights
• Best for: Regular job hunting, building a job database, market trend analysis

🎯 How to Use

Basic Usage (Recommended to start):

# Edit quick_job_research.py and change these lines:
JOB_ROLE = "Python Developer"  # Your target job
LOCATION = "San Francisco"     # Your preferred location

# Then run:
python quick_job_research.py

Advanced Usage:

# Edit simple_job_search.py for multi-platform search:
JOB_TITLE = "Full Stack Developer"
LOCATION = "Remote"
MAX_JOBS_PER_SITE = 10

python simple_job_search.py

📊 What You'll Get

Keyword Analysis:

• Most in-demand programming languages
• Popular frameworks and tools
• Essential soft skills
• Trending technologies

Market Insights:

• Salary ranges
• Experience level requirements
• Remote work availability
• Top hiring companies

Actionable Recommendations:

• Skills to prioritize learning
• Resume keyword optimization
• Best job boards for your field
• Profile improvement suggestions

🔧 Customization

Target Different Roles:

• "Data Scientist"
• "DevOps Engineer"
• "Product Manager"
• "UX Designer"
• "Machine Learning Engineer"

Location Options:

• "Remote"
• "New York, NY"
• "San Francisco, CA"
• "London, UK"
• "" (anywhere)

Platform Selection:

• LinkedIn Jobs
• Indeed
• AngelList/Wellfound
• Glassdoor
• Stack Overflow Jobs

🛠️ Troubleshooting

Common Issues:

1. Rate limiting: Add delays between requests
2. Captcha protection: Some sites may block automation
3. API limits: Google Gemini has usage quotas

Solutions:

• Use VPN if blocked
• Reduce MAX_JOBS number
• Add time.sleep() between searches
• Check your API key validity

📈 Advanced Features

Save Results:

# Results are automatically saved as JSON files
# Format: job_analysis_YYYYMMDD_HHMMSS.json

Schedule Regular Searches:

# Add to cron job for daily/weekly searches
0 9 * * 1 cd /path/to/project && python quick_job_research.py

Custom Analysis:

# Modify the analysis prompts to focus on:
# - Specific technologies
# - Company types
# - Salary negotiations
# - Career progression paths

🎯 Tips for Best Results

1. Be Specific: Use exact job titles like "Senior React Developer" vs "Developer"
2. Check Multiple Locations: Remote jobs often have more opportunities
3. Regular Updates: Job markets change quickly, run searches weekly
4. Keyword Optimization: Use the extracted keywords in your resume and applications
5. Company Research: Follow up on interesting companies found in results

🤝 Contributing

Feel free to:

• Add new job platforms
• Improve analysis prompts
• Add export formats (CSV, Excel)
• Create visualization dashboards
• Add salary prediction models

---