BhoomiAI

An AI-powered geospatial analysis platform for infrastructure planning, environmental assessment, and geographic intelligence.

🔎Overview

BhoomiAI leverages advanced AI and geospatial technology to provide actionable insights about locations and development opportunities. By combining data from OpenStreetMap, Open-Meteo, Wikipedia, and Census datasets with both Retrieval-Augmented Generation (RAG) and an Agentic AI pipeline, BhoomiAI enables intelligent geographic analysis through an intuitive LLM-based interface.

The platform integrates a tool-based reasoning agent that dynamically fetches and processes real-time geospatial data, allowing more accurate, contextual, and explainable insights for decision-making.

Key Capabilities

• Location Analysis - Explore detailed geographic and infrastructure data for any location
• Infrastructure Mapping - Analyze nearby infrastructure within a 15 km radius
• Demographic Insights - Access population statistics and development metrics
• AI-Powered Chatbot - Ask questions about construction, geography, and development opportunities
• Agentic AI Reasoning - Uses a tool-based AI agent to fetch, analyze, and reason over live geospatial data
• Data Aggregation - Unified access to multiple data sources through a single platform

---

🛠️Tech Stack

Layer	Technologies
Frontend	HTML, CSS, JavaScript
Backend	Python, LangChain, RAG, Agentic AI
LLM	Ollama (Local Inference - Qwen, Gemma)
Data Sources	OpenStreetMap, Open-Meteo, Wikipedia, Census Data

---

🤖Agentic Architecture

BhoomiAI incorporates an Agentic AI system implemented in a dedicated agent.py module.

• Uses tool-based execution for structured data retrieval
• Integrates multiple domain-specific tools (climate, infrastructure, demographics)
• Powered by local LLMs via Ollama (e.g., Qwen3, Gemma3)
• Enables multi-step reasoning instead of static responses
• Provides context-aware answers based on real-time location data

This architecture enhances the system from a traditional RAG pipeline to a more dynamic and intelligent decision-support system.

---

🚀Getting Started

Prerequisites

• Python 3.8 or higher
• Ollama installed on your system
• Required Python packages (see requirements.txt)

Installation

1. Clone the repository:

git clone https://github.com/shamveelcrx321/BhoomiAI.git
cd BhoomiAI

2. Set up the LLM model:

ollama run gemma3:4b

Note: You can use a different model (e.g., qwen3:8b) by updating the model name in agent.py or app.py

3. Install dependencies:

pip install -r requirements.txt

4. Run the application:

python app.py

5. Open your browser and navigate to:

http://localhost:8000

---

📁Project Structure

BhoomiAI/
├── app.py                 
├── agent.py               
├── data_fetcher.py       
├── requirements.txt       
├── static/               
│   ├── css/
│   ├── js/
│   └── index.html
└── README.md

---

👥Team

BhoomiAI was created by Team Fringes - CSE Batch (S2)

👑Leadership

Midhun K M - Team Lead

👥Team Members

• Abiraj TM
• Shamveel C
• Devananda C
• Sarang K

---

🙏Acknowledgement

The team would like to express special appreciation to Midhun K M for leading the project, coordinating the development process, and guiding the team throughout the workshop and hackathon.

---

🤝Support

For questions or issues, please reach out to the team lead:
Midhun K M
midhunkm1294@outlook.com