🤖 AI HR Recruiter Bot

This project demonstrates how to build a proactive AI Recruiter that actually reads resumes, understands context, and filters candidates before they reach a human.

It moves beyond linear if/else logic, using Google Gemini 2.5 to conduct natural conversations and analyze technical backgrounds.

---

Key Features

• It Reads PDFs: The bot extracts text from PDF resumes, analyzes the candidate's tech stack against vacancy requirements, and gives instant, relevant feedback.
• It Browses Links: Candidate sent a link to their portfolio or LinkedIn? No problem. The bot parses external content using httpx + BeautifulSoup.
• It Remembers Context: Thanks to Redis, the bot doesn't forget who you are or what file you sent 5 minutes ago.
• It's Professional: The system prompt is engineered to act as a gatekeeper. It politely deflects salary/benefit questions ("Let's discuss this at the interview") and focuses on technical fit.
• It's Clean: Fully typed Python 3.12, modular architecture, and packaged with Docker & Poetry.

Tech Stack

• Core: Python 3.12, Aiogram 3 (Asyncio)
• AI Engine: Google Gemini 2.5 (via google-generativeai)
• Infrastructure: Docker, Docker Compose
• Storage: Redis (FSM state & caching)
• Tools: Poetry, Pydantic, Makefile, Pytest

---

Quick Start

Everything is automated via Makefile.

Prerequisites

1. Docker & Docker Compose.
2. Telegram Bot Token (@BotFather).
3. Google Gemini API Key (Google AI Studio).

1. Configure

Create a .env file in the root directory (copy from .env.example):

BOT_TOKEN=your_telegram_bot_token
GOOGLE_API_KEY=your_gemini_api_key
REDIS_HOST=redis
REDIS_PORT=6379

2. Run (Docker) — Recommended

One command to rule them all:

make docker-up

• View logs: make docker-logs
• Stop: make docker-down

3. Run (Local)

If you want to run it without Docker (requires local Redis):

make install
make run

Quality Control

# Run Unit & Integration tests
make test

# Check code style (Black + Flake8)
make check

Structure

.
├── app
│   ├── bot          # Telegram handlers & UI
│   ├── services     # Logic: AI, Parser,
│   ├── config.py    # Strict config validation
│   └── main.py      # Entry point 
├── tests            # Comprehensive testing
├── Dockerfile       # Optimized for Poetry
└── Makefile         # Shortcuts