AI-powered B2B lead qualification system that automatically enriches, scores, and routes sales leads using LangGraph and Groq LLM.

Automatically processes incoming sales leads through an intelligent workflow:

1. Enrichment - Fetches company data (size, revenue, industry) via Clearbit API
2. AI Scoring - Uses Groq LLM to score lead quality (0-100) based on ideal customer profile
3. Smart Routing - Routes hot leads to CRM + Slack, warm leads to nurture, cold leads to drip campaigns
4. Actions - Creates Salesforce records and sends real-time notifications

Processing time: ~3 seconds per lead

New Lead → Enrich → AI Score → Route → CRM/Notify → Complete

Tech Stack:

• Orchestration: LangGraph (stateful agent framework)
• LLM: Groq API (Llama 3.3 70B)
• Language: Python 3.9+

# Clone the repository
git clone https://github.com/abdul-ITexpert/sales-lead-qualification-agent.git
cd sales-lead-qualification-agent

# Install dependencies
pip install -r requirements.txt

# Set up environment variables
cp .env.example .env
# Edit .env and add your Groq API key

from src.agent import create_lead_qualification_graph, process_lead
from src.config import Config

# Set API key
Config.GROQ_API_KEY = "your_groq_key_here"

# Initialize agent
graph = create_lead_qualification_graph()

# Process a lead
lead_data = {
    "email": "[email protected]",
    "company": "Acme Corp",
    "first_name": "John",
    "last_name": "Doe",
    "source": "website_form"
}

result = process_lead(lead_data, graph)

print(f"Score: {result['scoring'].score}/100")
print(f"Tier: {result['scoring'].tier}")

{
  "email": "[email protected]",
  "company": "Enterprise Tech Solutions",
  "first_name": "Sarah",
  "last_name": "Johnson"
}

{
  "score": 92,
  "tier": "hot",
  "reasoning": "VP-level decision maker at mid-market enterprise",
  "recommended_action": "Immediate outreach - schedule demo within 24 hours"
}

Edit src/config.py to customize scoring criteria:

IDEAL_COMPANY_SIZE = (200, 1000)      # Employees
MIN_REVENUE = 10_000_000               # $10M
TARGET_SENIORITY = ["VP", "C-Level", "Director"]
HOT_THRESHOLD = 70                     # Score threshold

pytest tests/

python examples/flask_api.py

MIT License - see LICENSE file for details.

Your Name - [email protected]

Project Link: https://github.com/yourusername/sales-lead-qualification-agent