🏆 Hackathon Copilot AI

Hackathon Copilot AI is an AI-powered assistant built using LangChain, Gemini, and Streamlit that helps students, developers, and hackathon enthusiasts transform raw ideas into stronger, more competitive hackathon projects.

Instead of repeatedly writing long prompts to analyze every new idea, this application automates the process and acts as your personal hackathon mentor.

---

✨ Features

🧠 AI-Powered Idea Analysis

Provide your project idea and receive:

• Refined problem statements
• Improved solution descriptions
• Potential impact analysis

---

💻 Tech Stack Recommendations

Generates:

• Fast MVP tech stack
• Advanced scalable tech stack

---

📊 Project Evaluation

The AI evaluates:

• Uniqueness
• Buildability
• Innovation
• Market potential
• Hackathon winning potential

---

🔍 Deep Project Review

The assistant identifies:

• Hidden weaknesses
• Missing features
• Potential implementation challenges
• Improvement opportunities

---

🚀 Hackathon-Focused Guidance

Helps users:

• Strengthen ideas
• Improve project pitches
• Build more competitive hackathon solutions

---

🏗️ Architecture

User
   │
   ▼
Streamlit UI
   │
   ▼
Prompt Template
   │
   ▼
LangChain Chain
   │
   ▼
Gemini LLM
   │
   ▼
Output Parser
   │
   ▼
Final Structured Response

---

⚙️ How It Works

The application follows a simple LangChain workflow:

Step 1

User enters:

• Project idea
• Hackathon track
• Preferred tech stack

Step 2

A custom PromptTemplate structures the request.

Step 3

LangChain sends the prompt to Gemini.

Step 4

Gemini generates a detailed analysis.

Step 5

The Output Parser cleans the response and returns only the final text.

---

📂 Project Structure

hackathon-copilot-ai/
│
├── app.py                # Streamlit frontend
├── chains.py             # LangChain pipeline
├── prompts.py            # Prompt templates
├── requirements.txt      # Dependencies
├── .env                  # API keys
├── README.md             # Project documentation
│
└── assets/
    └── screenshots/

---

🛠️ Tech Stack

• Python
• Streamlit
• LangChain
• Gemini API
• Prompt Engineering

---

🔑 Environment Variables

Create a .env file in the root directory.

GOOGLE_API_KEY=your_google_api_key

---

📦 Installation

Clone Repository

git clone https://github.com/yourusername/hackathon-copilot-ai.git

cd hackathon-copilot-ai

Create Virtual Environment

Windows

python -m venv venv

venv\Scripts\activate

Linux/Mac

python3 -m venv venv

source venv/bin/activate

---

Install Dependencies

pip install -r requirements.txt

---

▶️ Run Application

streamlit run app.py

Application will run on:

http://localhost:8501

---

📝 Example Usage

Input

Project Idea:
AI-powered disaster communication system

Hackathon Track:
Social Impact

Preferred Tech Stack:
Python, FastAPI, React, ESP32

---

Sample Output

Problem:
Communication networks often fail during disasters.

Solution:
Create a decentralized LoRa-based emergency
mesh communication system powered by AI.

Impact:
Enables communication even when mobile towers
are unavailable.

Tech Stack Option 1:
Python + FastAPI + Streamlit + ESP32

Tech Stack Option 2:
FastAPI + React + PostgreSQL + MQTT + ESP32

Uniqueness Score:
8.5/10

Buildability Score:
9/10

---

📸 Screenshots

hack_copilot.png

---

🔮 Future Improvements

• Web Search Integration
• Research Paper Search
• Multi-Agent Workflows
• Memory Support
• Hackathon Winner Comparison
• GitHub Repository Search
• PDF Report Generation
• Pitch Deck Generation
• LangGraph Integration

---

🤝 Contributing

Contributions, feature requests, and suggestions are welcome.

Feel free to fork this repository and create a pull request.

---

📌 Learning Outcomes

This project helped me learn:

• LangChain fundamentals
• Prompt Engineering
• LLM pipelines
• Streamlit application development
• Environment variable management
• Building AI applications from scratch

---

🏷️ Tags

LangChain
Gemini
LLM
Generative AI
Streamlit
Python
Prompt Engineering
Hackathon
Artificial Intelligence
Machine Learning

---

👩‍💻 Author

Thivisha

Built with ❤️ while exploring LangChain and Agentic AI.