🧠 A multi-agent Gen AI platform powered by Google Gemini 2.5 Pro that autonomously generates, reviews, and composes full-length academic research papers — complete with chat assistant, dark UI, and editable .docx export.

🧠 AI Research Lab Simulator

"Empowering the next generation of researchers — one AI agent at a time."

🚀 Overview

The AI Research Lab Simulator is a next-generation multi-agent Gen AI platform powered by Google Gemini 2.5 Pro, designed to autonomously generate, review, and finalize research papers. It replicates the roles of human researchers — from ideation to publication.

View my live project here: PROJECT

⚠️ IMPORTANT NOTE
When you open the live demo link, you might see a message saying:
“This app is sleeping due to inactivity.”

⏰ Please click the “Wake up” button and wait for 30–60 seconds.
Streamlit Cloud will automatically restart the app — after which everything will work perfectly! 🚀

💡 (This is a normal behavior for free-tier hosting and doesn’t affect functionality.)

🧩 Features

✅ Multi-Agent Workflow — 7 AI agents simulate the human research process.
✅ Real-time Streamlit Interface with glassy, futuristic design.
✅ Interactive AI Chat with persistent context memory.
✅ One-click .docx export for editable research papers.
✅ Secure API handling with .env integration.
✅ Fully dark, modern UI optimized for laptops and tablets.

🧠 Agent Architecture

Agent	Role	Description
🧠 Researcher	Content Creator	Generates a detailed draft on the given topic.
🔍 Reviewer	Logic Analyst	Refines structure and logical accuracy.
✍️ Editor	Language Expert	Polishes grammar, tone, and readability.
🧩 Citation Builder	Reference Curator	Adds citations and references.
🧪 Fact Verifier	Accuracy Specialist	Ensures factual correctness.
🧭 Summarizer	Synthesizer	Produces a 200-word executive summary.
📚 Research Composer	Integrator	Combines all outputs into one cohesive paper.

🏗️ Tech Stack

Language: Python 3.10+
Framework: Streamlit
Model: Google Gemini 2.5 Pro API
Libraries: google-generativeai, python-docx, python-dotenv, streamlit
Output Format: .docx

⚙️ Installation & Setup

git clone https://github.com/Pranav-Sharma-Official/AI-Research-Lab-Simulator.git
cd AI-Research-Lab-Simulator
pip install -r requirements.txt

Create a .env file:

GEMINI_API_KEY=your_gemini_api_key_here

▶️ Running the Application

streamlit run app.py

Steps:

Enter your Gemini API Key and Research Topic.
Click Generate Full Research Paper.
Watch each agent work in real-time.
Chat with your AI assistant for clarifications.
Export the final .docx research paper.

🖥️ UI & UX Highlights

UI Screenshot
Flowchart

🖤 Dark glassy UI with glowing cyan elements.
📊 Live progress tracking for all 7 agents.
💬 Persistent chat assistant using Streamlit’s chat UI.
🧾 Real-time result rendering with collapsible outputs.
💡 Responsive for both laptops and tablets.

💡 Innovation & Impact

This project bridges AI and Academia, demonstrating how Gen AI can assist in research content generation.
It reduces manual drafting time, enhances accuracy, and supports researchers, students, and educators in academic publishing.

Applications:

Thesis & Dissertation Drafting
Automated Whitepaper Generation
Education & Research Training
Scientific Paper Ideation

📊 Evaluation Metrics

Criteria	Description	Status
Functionality	7 agents executed sequentially with distinct roles	✅
Innovation	Multi-agent AI workflow + real-time chat	✅
Design	Glassmorphic UI with modern visuals	✅
Usability	Editable `.docx` export & persistent chat	✅
Scalability	Modular architecture for domain-specific research	✅

⚠️ Known Issues & Fixes

Issue	Description	Resolution
PDF Encoding	FPDF couldn’t handle emojis	Switched to `.docx` output
Streamlit Chat Reset	Session re-runs lost chat state	Fixed via `st.session_state`
API Latency	Gemini long response time	Added spinner + error handling

🔮 Future Scope

🔗 Multi-user collaborative editing mode.
🎙️ Voice assistant for accessibility.
📡 Cloud integration (Google Drive / GitHub save).
🔍 Built-in plagiarism detection.
🧾 Auto reference verification and citation fetching.

👥 Team & Contributions

Member	Role	Contributions
Pranav Sharma	Team Lead & Developer	Project architecture, backend, prompt engineering, Testing, documentation, and feature QA.

🏆 Why This Project Wins

✨ End-to-end AI automation for research generation.
✨ Functional + visually stunning interface.
✨ Real-world applicability with Gemini 2.5 Pro.
✨ Demonstrates deep technical and design skills.
✨ Complete, ready-to-show hackathon-grade product.

AI-Research-Lab-Simulator