AI-Powered Resume Analyzer & RAG Chatbot

A production-ready backend application built with Spring Boot, Spring AI, VectorStore, and Apache Tika. This project provides APIs for:

• Resume text extraction
• AI-based resume analysis
• ATS score calculation
• RAG-powered PDF-based chatbot
• General AI chat interface

The system integrates AI with traditional backend engineering to create a powerful resume evaluation and document-aware chatbot.

---

🚀 Features

1. Resume Analyzer (PDF → Text → AI Evaluation)

• Extracts resume text using Apache Tika
• Identifies key skills, improvements, and ratings
• Returns clean, structured JSON responses

2. ATS Checker (Resume vs Job Description)

• Compares uploaded resume with provided job description
• Generates an ATS score (0–100)
• Lists matched & missing keywords
• Provides a short summary for improvement

3. RAG Chatbot (Context-Aware Document QA)

• Uses SimpleVectorStore to store embeddings
• Answers questions strictly from PDF content
• Uses QuestionAnswerAdvisor + semantic search

4. General AI Chatbot

• Pure LLM responses using Spring AI
• Stateless, fast, and simple

---

📌 Tech Stack

• Java 21
• Spring Boot 3.5+
• Spring AI (ChatClient, Embeddings, Vector Store, Advisors)
• Apache Tika (Document parsing)
• SimpleVectorStore (JSON-based vector DB)
• Spring Security (optional)
• REST APIs
• Postman (API testing)

---

📁 Project Structure

src/main/java/com/.../controller
    └── ChatController.java

src/main/java/com/.../config
    ├── ChatConfig.java
    ├── WebConfig.java

src/main/java/com/.../advisor
    └── LoggingAdvisor.java

src/main/resources
    ├── input-data/        # PDF files for RAG
    └── output-data/       # vectorstore.json

---

🔗 API Endpoints

1. Health Check

GET /api/health

2. General Chatbot

POST /api/chat
{
  "message": "Hello"
}

3. Direct Question

GET /api/ask?question=What+is+Spring+AI

4. RAG Chatbot

POST /api/rag/chat
{
  "message": "What is mentioned about microservices in the PDF?"
}

5. Resume Analyzer

POST /api/analyze
Content-Type: multipart/form-data
file = <resume.pdf>

6. ATS Checker

POST /api/ats-check
Content-Type: multipart/form-data
file = <resume.pdf>
jd = "Your Job Description..."

---

📦 Vector Store (RAG)

The application automatically:

1. Reads PDFs from src/main/resources/input-data/
2. Extracts text using Tika
3. Splits into chunks using TokenTextSplitter
4. Generates embeddings
5. Saves vector DB as vectorstore.json

This ensures fast, accurate document-based QA.

---

🧪 Testing with Postman

Each endpoint supports manual testing through Postman:

• Use form-data for file uploads
• Use raw JSON for chat endpoints
• Verify vector store by calling RAG endpoint

---

📘 How to Run

1. Clone the project

git clone <repo-url>
cd project-folder

2. Configure application properties

Add your OpenAI / API provider keys.

3. Run the application

mvn spring-boot:run

4. Test APIs using Postman or cURL.

---

🌟 Why This Project Is Valuable

This backend demonstrates:

• AI + backend integration
• RAG architecture in production
• PDF processing & embeddings
• API design & error handling
• Vector databases in Spring Boot

Perfect for showcasing backend + AI skills on your resume.

---

✨ Author

Chandan Dhabele Java Backend Developer | Spring Boot | Microservices | AI Integrations

---