Sourcing Agent 🔍

Welcome to Arpita's Autonomous Sourcing Agent – a powerful, fast, and Gen Z-friendly tool that searches, scores, and connects ML talent with top-tier opportunities.

⚙️ Powered by Python + Streamlit + SerpAPI 🔗 Live Demo: [https://youtu.be/YDY5Q3Fs8PY] 🧑‍💼 Job Role: ML Research Engineer @ Windsurf (Codeium)

---

🚀 Features

✅ Job-Aware Search

Parses a detailed JD and crafts smart Google Search queries to find real LinkedIn profiles.

✅ Fit Scoring Algorithm

Calculates fit scores (1–10) based on: Education, Career Trajectory, Company, Skills, Location, and Tenure.

✅ AI-Powered Outreach Generator

GPT-driven, personalized LinkedIn messages tailored to each candidate's strengths.

✅ Interactive Streamlit UI

Clean, wide-screen view to explore top candidates, view breakdowns, and copy messages instantly.

✅ Export to CSV + HTML

Neat outputs for recruiters to share, store, and analyze.

---

📸 Demo Video (Loom) – Coming Soon

You'll see:

• 📥 A job description being parsed
• 🔗 LinkedIn profiles discovered
• 📊 Fit scores calculated
• 💌 Outreach messages generated
• 🖥️ Streamlit app in action!

---

🧪 How to Run Locally

1. Clone the Repo

git clone https://github.com/Jexa07/sourcing-agent.git
cd sourcing-agent

2. Install Dependencies

pip install -r requirements.txt

3. Add Your API Key

Create a .env file in the root folder:

SERPAPI_KEY=your-serpapi-key-here

4. Run the App

streamlit run app.py

---

📂 Project Structure

.
├── app.py                # Streamlit frontend
├── main.py               # End-to-end pipeline script
├── core/                 # Logic modules: scraping, scoring, parsing
├── utils/                # Logger, helpers
├── data/                 # Inputs & outputs
├── .env                  # Your API key (not committed)
├── requirements.txt      # Dependencies
└── README.md             # This file ❤️

---

📊 Sample Output

{
  "name": "Jane Smith",
  "linkedin_url": "https://linkedin.com/in/janesmith",
  "company": "OpenAI",
  "fit_score": 9.2,
  "score_breakdown": {
    "education": 9.5,
    "trajectory": 9.0,
    "company": 9.0,
    "skills": 10.0,
    "location": 8.0,
    "tenure": 9.0
  },
  "outreach_message": "Hi Jane, I came across your profile..."
}

---

💡 Bonus Ideas (Planned)

• Confidence Scoring when data is incomplete
• GitHub/Twitter enhancement for deeper signals
• FastAPI endpoint for submission API
• Batch job support (10+ jobs)

---

👋 About Me

Hey! I'm Arpita Pani, a CSE student with a passion for intelligent systems and sleek UX. This project was an absolute thrill to build — reach out if you'd like to connect!

🔗 LinkedIn | 🐙 GitHub

---

📜 License

MIT License – free to use, remix, and level up the future of sourcing.