recruit-ai-agent

This project is an AI-powered recruitment workflow using Mastra AI agents to automate initial email screening, validate candidate submissions (resume/cover letter), and send appropriate responses. The system is designed to be modular, context-aware, and easily extendable.

Tech Stack

• Mastra AI
• Node.js
• Redis
• BullMQ
• Gmail API
• TypeScript

Installation

1. Clone the repository using git clone git@github.com:togetherai/recruit-ai-agent.git
2. Install the dependencies using npm install or yarn install
3. Set up your Gmail account with the necessary permissions using the OAuth2.0 method.
4. Create a .env file with the following variables:
   • RECRUITMENT_MAIL: The Gmail email address to be used for the recruitment workflow
   • REDISCLOUD_URL: The Redis Cloud URL to be used for storing the job application data
   • GMAIL_CLIENT_ID: The Gmail API client ID
   • GMAIL_CLIENT_SECRET: The Gmail API client secret . Create a job application database in your Redis Cloud instance.
5. Run the application using npm run start or yarn start
6. The application will start the Mastra AI agent and the BullMQ job queue.

Usage

1. Send an email to the RECRUITMENT_MAIL address with the job application details.
2. The system will automatically create a job application database in your Redis Cloud instance.
3. The system will then validate the candidate submissions (resume/cover letter) and send appropriate responses.

Configuration

The system can be configured using the following environment variables:

• RECRUITMENT_MAIL: The Gmail email address to be used for the recruitment workflow
• REDISCLOUD_URL: The Redis Cloud URL to be used for storing the job application data
• GMAIL_CLIENT_ID: The Gmail API client ID
• GMAIL_CLIENT_SECRET: The Gmail API client secret
• BULLMQ_REDIS_URL: The Redis URL to be used for the BullMQ job queue
• BULLMQ_CONCURRENCY: The number of concurrent jobs to be processed by the BullMQ job queue
• BULLMQ_REDIS_KEY_PREFIX: The key prefix to be used for the BullMQ job queue in Redis