ScholAR: Advanced Research Assistant with Retrieval

🧠 Overview

ScholAR is an AI-powered research assistant built as a Kalvium capstone project. It combines RAG, multi-persona prompting, and chain-of-thought reasoning to help researchers and students manage complex research workflows.

✨ Key Features

🎭 Multi-Persona AI System

• Literature Reviewer: Comprehensive paper analysis and synthesis
• Data Analyst: Statistical interpretation and visualization
• Citation Manager: Automated formatting and validation
• Writing Assistant: Academic writing support and editing

🎛️ Dynamic Parameter Tuning

• Automatic temperature adjustment based on task type
• Context-aware response length optimization
• Domain-specific parameter profiles

📊 Structured Output Generation

• Standardized research summaries (Abstract, Findings, Methodology)
• Multi-format citations (APA, MLA, Chicago, IEEE)
• JSON metadata for reference managers

🧩 Retrieval-Augmented Generation (RAG)

• Personal knowledge base creation and management
• Cross-document intelligent querying
• Automatic knowledge linking and synthesis

🚀 Quick Start

Prerequisites

• Node.js 16+
• An LLM API key (OpenAI or compatible)

Installation

1. Clone the repository

git clone https://github.com/Uday9909/scholAR.git
cd scholAR

2. Install dependencies

npm install

3. Configure environment variables

cp .env.example .env
# Add your API keys to .env

4. Start the app

npm run dev

📖 Usage Examples

Basic Research Query

const response = await assistant.query({
  question: "What are recent developments in quantum computing?",
  persona: "literature_reviewer",
  output_format: "structured_summary"
});

RAG Query

const insights = await assistant.ragQuery({
  question: "How do quantum algorithms compare to classical ones?",
  sources: ["personal_library"]
});

🏗️ Architecture

┌─────────────────┐    ┌──────────────────┐    ┌─────────────────┐
│   Frontend      │    │   API Gateway    │    │   AI Engine     │
│   (React)       │◄──►│   (Express)      │◄──►│   (LangChain)   │
└─────────────────┘    └──────────────────┘    └─────────────────┘
                                │                        │
                       ┌──────────────────┐    ┌─────────────────┐
                       │   Vector DB      │    │   Function      │
                       │   (Pinecone)     │    │   Registry      │
                       └──────────────────┘    └─────────────────┘

🔧 Configuration

System Prompts

Located in config/prompts/, customize AI personas:

literature_reviewer:
  system_prompt: "You are an expert academic literature reviewer..."
  temperature: 0.3
  max_tokens: 2000

RAG Settings

embedding_model: "text-embedding-ada-002"
chunk_size: 1000
overlap: 200
similarity_threshold: 0.75

🤝 Contributing

1. Fork the repository
2. Create a feature branch (git checkout -b feature/your-feature)
3. Commit your changes (git commit -m 'Add your feature')
4. Push to the branch (git push origin feature/your-feature)
5. Open a Pull Request

🙏 Acknowledgments

• LangChain for the AI application framework
• OpenAI for GPT models and embeddings

---

Built by Udaybir Singh