RAG Microservice - Professional Profile Assistant

The RAG Microservice is an intelligent assistant designed to provide detailed, structured, and professional responses about Kushagra Sikka's professional profile. It uses Retrieval-Augmented Generation (RAG) to retrieve relevant information and generate well-formatted, concise answers tailored for recruiters and collaborators.

Use Case

This system is tailored to Kushagra's recruiters and professional network, allowing them to:

• Fetch structured professional summaries.
• Retrieve details about technical skills with usage examples.
• Explore academic achievements, work experience, and projects.
• Gain insights into recent contributions and research focus.
• Access contact information and links to professional profiles.

The assistant ensures:

• Accurate and verified information retrieval from curated documents.
• Well-formatted responses using bullet points for readability.
• A professional tone to enhance user experience and utility.

Example Questions

• "Who is Kushagra Sikka?"
  Provides a professional overview, recent impact, and technical expertise.

• "What are Kushagra's technical skills?"
  Details programming languages, cloud platforms, and tools with usage examples.

• "Tell me about Kushagra's achievements."
  Highlights teaching impact, technical accomplishments, and academic recognition.

---

Architecture

The system integrates:

• Backend: FastAPI + Haystack for RAG implementation.
• Frontend: React + TailwindCSS for a user-friendly interface.
• Deployment: Docker, Jenkins, and AWS EC2 for production readiness.

Backend Workflow

1. Document Store: Stores structured professional data (e.g., skills, projects, achievements).
2. Retriever: Fetches relevant documents based on the query.
3. Prompt Builder: Constructs dynamic prompts tailored to the query type.
4. Generator: Uses a text generation model to produce well-formatted, bullet-pointed responses.

Frontend Features

• Interactive UI for querying Kushagra's profile.
• Real-time response rendering with React.
• Mobile-responsive design using TailwindCSS.

---

Local Development

Follow these steps to set up the project locally:

1. Clone the Repository

git clone https://github.com/YourUsername/RAG_Microservice.git
cd RAG_Microservice

2. Set Up the Backend

python -m venv venv
source venv/bin/activate  # On Windows use: .\venv\Scripts\activate
pip install -r requirements.txt

3. Set Up the Frontend

cd rag-frontend
npm install

4. Create Environment Variables

Copy the example .env file and update it with your configuration:

cp .env.example .env

5. Run the Application Locally

Backend

uvicorn rag_microservice.app:app --reload

Frontend

In a new terminal:

cd rag-frontend
npm start

---

Docker Deployment

To deploy the application using Docker:

docker-compose up -d

---

Production Deployment

1. Configure an AWS EC2 instance with the necessary environment.
2. Set up Docker and Jenkins for continuous integration and deployment.
3. Refer to the deployment/README.md file for step-by-step instructions.

---

Project Structure

RAG_Microservice/
├── rag_microservice/         # Backend code
│   └── app.py                # Main FastAPI application
├── rag-frontend/             # React frontend
├── data/                     # Data directory
├── docker-compose.yml        # Docker compose configuration
├── Jenkinsfile               # CI/CD pipeline
└── deployment/               # Deployment documentation and scripts

---

Environment Variables

Backend

Variable	Description
CORPUS_DOCUMENTS_PATH	Path to the document corpus (e.g., professional details).
TEXT_EMBEDDING_MODEL	Model for text embeddings (e.g., sentence-transformers/all-MiniLM-L6-v2).
GENERATOR_MODEL	Model for text generation (e.g., google/flan-t5-large).

Frontend

Variable	Description
REACT_APP_API_URL	Backend API URL (e.g., http://localhost:8000).

---

Features

Backend

• Implements RAG using Haystack components:
  • Document Splitter: Chunks documents for better retrieval.
  • Retriever: Retrieves relevant documents based on queries.
  • Prompt Builder: Constructs dynamic prompts for generation.
  • Generator: Produces well-structured responses.
• Pre-processed corpus includes:
  • Professional profile, achievements, and technical skills.
  • Work experience and key projects.
  • Contact information.

Frontend

• Query interface with a clean, professional design.
• Supports real-time query results.
• Mobile-friendly and intuitive layout.

---

How It Works

1. Ask a Question: The user asks a question, such as "What are Kushagra's skills?".
2. Document Retrieval: The system retrieves relevant sections from the document store.
3. Dynamic Prompt Creation: A prompt is generated based on the question type.
4. Answer Generation: The model generates a structured, bullet-pointed response.
5. Response Display: The frontend displays the response in a user-friendly format.

---

Use Case for Recruiters

Objective

To provide a quick, reliable, and structured way for recruiters to:

• Understand Kushagra's professional profile.
• Explore technical expertise and projects.
• Gain insights into recent achievements and research focus.

Advantages

• Time-Saving: Direct answers without the need to parse lengthy resumes.
• Structured Format: Responses are concise and categorized for better readability.
• Accurate Data: Fetches verified information only.

Example Queries

• "Who is Kushagra Sikka?"
  • Learn about Kushagra's current roles, educational background, and recent impact.
• "What are Kushagra's achievements?"
  • Understand his teaching impact, technical accomplishments, and academic recognition.
• "Tell me about Kushagra's skills."
  • Explore his technical expertise with usage examples.

---

Contributing

1. Fork the repository.
2. Create a feature branch:

   git checkout -b feature/AmazingFeature

3. Commit your changes:

   git commit -m 'Add AmazingFeature'

4. Push to the branch:

   git push origin feature/AmazingFeature

5. Open a Pull Request.

---

Contact

For further queries or contributions, reach out to Kushagra Sikka:

• Email: [email protected]
• Portfolio: kushagrasikka.com
• GitHub: github.com/KushagraSikka
• LinkedIn: linkedin.com/in/kushagrasikka