A sophisticated multi-agent recruitment pipeline built with LangGraph, Google Gemini 2.5 Flash, and Streamlit. HireGenAI automates talent acquisition by parsing resumes, analyzing job descriptions, and generating personalized interview strategies.
- 📄 Resume Parser — Extracts structured data (Skills, Experience, Education) from PDF/TXT files.
- 🎯 JD Analyzer — Breaks down job descriptions into required skills and minimum experience.
- ⚖️ Hybrid Scorer — Ranks candidates using a weighted formula:
$60%$ Skills +$30%$ Experience +$10%$ Education. - 💡 AI Insights — Generates 3 concise bullet points explaining why a candidate stands out.
- 💬 Interview Architect — Crafts 5 tailored technical questions based on the candidate's background.
| Tool | Purpose |
|---|---|
| Streamlit | Web UI & Dashboard |
| LangGraph | Multi-agent state machine orchestration |
| Google Gemini 2.5 Flash | LLM powering extraction and reasoning |
| Pydantic | Structured data validation |
| PyPDF2 | PDF text extraction engine |
The system uses a sequential StateGraph to ensure data flows accurately:
[Resume & JD Input] │ ▼ [Resume Parser Node] ──► Structured JSON Data │ ▼ [JD Analyzer Node] ──► Extraction of Requirements │ ▼ [Matching Agent] ──► Calculates Match & Missing Skills │ ▼ [Interview Agent] ──► Generates 5 Custom Questions
HireGenAI/ │ ├── app.py # Main application & LangGraph logic ├── requirements.txt # Python dependencies └── outputs/ # Application screenshots ├── 1.jpeg
├── 2.jpeg
└── 3.jpeg
This project is open-source and available under the MIT License.