AI-driven candidate evaluation

A .NET 8 Web API that analyzes CVs/resumes against job descriptions using Azure OpenAI to provide skill gap analysis and ATS scoring.

Features

• 📄 Multi-format Document Support: Supports PDF (.pdf) and Word (.docx) file formats
• 🤖 AI-Powered Analysis: Uses Azure OpenAI GPT-4.1 for intelligent CV analysis
• 📊 ATS Scoring: Provides Applicant Tracking System compatibility scores (0-100)
• 🔍 Skill Gap Analysis: Identifies missing or weak skills compared to job requirements
• 📝 JSON Response: Returns structured analysis results in JSON format
• 🛡️ Error Handling: Comprehensive error handling and logging

Prerequisites

• .NET 8 SDK
• Azure OpenAI Service account with GPT-4.1 model deployment
• Visual Studio 2022 or VS Code (recommended)

Installation

1. Clone the repository

2. Install Required NuGet Packages

dotnet add package Azure.AI.OpenAI
dotnet add package DocumentFormat.OpenXml
dotnet add package PdfPig

Or add these to your .csproj file:

<PackageReference Include="Azure.AI.OpenAI" Version="2.1.0" />
<PackageReference Include="DocumentFormat.OpenXml" Version="3.1.0" />
<PackageReference Include="PdfPig" Version="0.1.8" />

3. Configure Azure OpenAI

Edit Program.cs and replace the placeholder values:

var endpoint = "xxxxx";
var apiKey = "xxxxx";

Ensure you have a GPT-4.1 model deployed named gpt-4.1 or just change the name of the deployment model in the program.cs to your available one.

4. Run the Application

cd ./CvJdSkillGapApi/
dotnet run

The API will be available at http://localhost:5161 (or the port shown in console output).

API Usage

Endpoint

POST /analyze

Request Format

Send a multipart/form-data request with:

Field	Type	Required	Description
cv	File	Yes	CV/Resume file (.pdf or .docx)
job_description	Text	Yes	Job description text

Example Request (cURL)

curl -X POST http://localhost:5161/analyze \
  -F "cv=@resume.pdf" \
  -F "job_description=We are looking for a Senior Software Developer with experience in C#, .NET, Azure, and microservices architecture..."

Example Request (Postman)

1. Set method to POST
2. URL: http://localhost:5161/analyze
3. Go to Body → form-data
4. Add key cv (type: File) and select your CV file
5. Add key job_description (type: Text) and paste the job description

Response Format

{
  "skillGaps": [
    "Machine Learning frameworks (TensorFlow, PyTorch)",
    "Kubernetes orchestration",
    "Advanced DevOps practices"
  ],
  "atsScore": 75
}

Response Fields

• skillGaps: Array of strings listing missing or weak skills
• atsScore: Integer from 0-100 indicating ATS compatibility

Supported File Formats

• PDF (.pdf): Uses PdfPig library for accurate text extraction
• Word (.docx): Uses DocumentFormat.OpenXml for text extraction

Development

Project Structure

├── Program.cs              # Main application entry point
├── CV-JD-Analyzer.csproj   # Project file with dependencies
└── README.md              # This file

Key Components

• File Processing: Handles PDF and DOCX text extraction
• Azure OpenAI Integration: Manages API calls to GPT-4.1
• Error Handling: Comprehensive exception handling and logging
• Validation: Input validation and null checks

Common Issues

1. "Could not extract text from resume"

   • Ensure the PDF/DOCX file is not corrupted
   • Check if the file contains extractable text (not just images)
   • Verify file size is reasonable (< 10MB recommended)

2. Azure OpenAI Connection Issues

   • Verify your API key and endpoint URL
   • Ensure your Azure OpenAI resource is active
   • Check that GPT-4 model is deployed and accessible

3. File Upload Issues

   • Ensure you're sending multipart/form-data
   • Check file field name is exactly cv
   • Verify job description field name is exactly job_description

License

This project is licensed under the MIT License.

---

Note: This is a backend API only. You'll need to create a frontend application or use tools like Postman/cURL to interact with the API.