Recruitt - AI Talent Screening Platform

Recruitt is a powerful, high-performance talent acquisition engine designed to optimize the intelligence matrix for modern recruiters. It leverages Gemini 2.5 Flash for deep semantic analysis of resumes and job requirements.

Local Development

To get the full Recruitt engine running on your machine, follow these steps:

1. Prerequisites

• Node.js: v18.0.0 or higher
• pnpm: Fast, disk-space-efficient package manager
• MongoDB: A running instance (local or Atlas)
• Gemini API Key: Required for the AI screening pipeline

2. Installation

# Install dependencies for the entire monorepo
pnpm install

3. Environment Configuration

Create an .env file in apps/server/.env:

PORT=5000
MONGODB_URI=your_mongodb_connection_string
GEMINI_API_KEY=your_gemini_api_key
RESEND_API_KEY=your_resend_api_key
EMAIL_FROM="Recruitt <notifications@yourdomain.com>"

Create an .env.local file in apps/web/.env.local:

NEXT_PUBLIC_API_URL=http://localhost:5000/api
AUTH_SECRET=your_next_auth_secret
PRELAUNCH_PASSCODE=your_preferred_passcode
DISABLE_PASSCODE_SCREEN=false

4. Running the Platform

You can start all services (Frontend, Backend, and Docs) simultaneously from the root:

pnpm dev

• Dashboard: http://localhost:3000
• Server: http://localhost:5000
• Docs: http://localhost:3001

---

Technical Architecture

Recruiter User Flow

The frontend is designed for high-velocity decision making, focusing on clarity and AI explainability.

journey
    title Recruiter Journey
    section Job Setup
      Create Job Posting: 5: Recruiter
      Define Requirements: 4: Recruiter
    section Talent Ingestion
      Upload PDF Resumes: 5: AI-Parser
      Bulk Import CSV/Excel: 5: AI-Mapper
    section Matrix Evaluation
      Adjust Screening Weights: 4: Recruiter
      Trigger Matrix Assessment: 5: Gemini-AI
    section Final Selection
      Review Leaderboard: 5: Recruiter
      Shortlist Candidates: 5: Recruiter

Loading

AI Decision Pipeline

The backend orchestrates semantic comparisons between talent profiles and job specifications using a weighted matrix approach.

graph TD
    A[Recruiter Triggers Screening] --> B[Fetch Job & Applicant Data]
    B --> C[Flatten Data for LLM Context]
    C --> D[Apply User Weights: Skills/Exp/Edu]
    D --> E[Gemini 2.5 Flash Processing]
    E --> F[Generate JSON Output]
    F --> G[Extract Strengths/Gaps/Scores]
    G --> H[Persist to MongoDB]
    H --> I[Notify Frontend Leaderboard]

Loading

---

Sub-Project Documentation

Explore more detailed documentation for each part of the platform:

• 🎨 Frontend Dashboard - Next.js 16 (App Router) + Tailwind 4
• ⚙️ Backend Orchestrator - Node.js + TypeScript + Mongoose
• 📚 Overall Documenation - Built with Docus
• 📦 Shared Library - Types and Talent Profile Specifications

---

© 2026 Recruitt Technologies. Optimizing the future of talent.