AI Hiring Automation for Frappe/ERPNext

AI-powered recruitment automation that streamlines the entire hiring pipeline from resume screening to interview preparation.

🚀 Overview

AI Hiring Automation is a comprehensive Frappe application that leverages artificial intelligence to automate and optimize the recruitment process. Built on the Frappe Framework with seamless ERPNext HRMS integration, it reduces time-to-hire while improving candidate quality through intelligent screening and evaluation.

✨ Key Features

• 📄 Intelligent Resume Parsing: Automatically extract structured data from PDF, DOCX, and TXT resumes
• 🎯 AI-Powered Candidate Shortlisting: Evaluate candidates against job requirements with detailed scoring
• ❓ Dynamic Questionnaire Generation: Create customized yes/no questionnaires tailored to each candidate
• 📊 Automated Answer Evaluation: Analyze responses with weighted scoring and topic-wise assessment
• 📝 Interview Brief Generation: Prepare comprehensive interview guides with suggested questions
• 🔄 Complete Workflow Management: Track candidates through 13 pipeline stages with state validation
• 📧 Automated Notifications: Send email notifications to candidates and HR at key milestones
• 📈 Analytics & Reporting: Track AI performance, pipeline metrics, and hiring funnel statistics
• 🔒 Enterprise Security: Rate limiting, audit logging, GDPR compliance, and role-based access control
• 🔌 REST API: Full-featured API for integrations and custom workflows

Requirements

• Frappe Framework v14 or higher
• Frappe HRMS
• OpenAI-compatible API endpoint (OpenAI, Azure OpenAI, or Ollama)
• Redis (for background jobs)
• Python 3.10+

Installation

1. Get the app:

bench get-app https://github.com/connectionloops/frappe_ai_hiring

2. Install on your site:

bench --site your-site.local install-app frappe_ai_hiring

3. Configure AI Settings:
   • Navigate to AI Settings in your site
   • Add your OpenAI-compatible API credentials
   • Configure model preferences and temperature

Usage

For HR Managers

1. Job Applicant Processing: When a new candidate applies with a resume, the system automatically:

   • Parses the resume
   • Evaluates against the job description
   • Provides shortlisting recommendations

2. Review AI Decisions: Navigate to AI Shortlisting Result to see:

   • Fit scores
   • Detailed reasoning
   • Missing skills analysis

Configuration

AI Settings

• Provider: OpenAI-compatible endpoint
• API Base URL: Your API endpoint
• API Key: Encrypted credentials
• Default Model: e.g., gpt-4-turbo-preview
• Temperature: 0.2 (recommended for consistency)
• Timeout: 60 seconds

Queue Configuration

Background jobs use Frappe's queue system:

• Queue: long (for LLM calls)
• Timeout: 300 seconds
• Retry Logic: 3 attempts with exponential backoff

Security & Compliance

• PII Redaction: Names, emails, phone numbers are redacted before LLM calls
• Data Encryption: Sensitive fields encrypted at rest
• Audit Logs: Every AI decision logged with prompt and response
• No Auto-Hire: System provides recommendations only; humans make final decisions

Pipeline States

Applied → AI Parsed → AI Shortlisted → Interview Scheduled → Interview Completed →
Offer / Rejected

License

MIT License - see LICENSE file for details

Credits

Built with ❤️ using Frappe Framework