AskAtlas AI Agent

Overview

AskAtlas is an intelligent AI agent designed to provide comprehensive insights into global population and demographic data. Leveraging advanced language models and query engines, AskAtlas can answer complex questions about world populations, demographics, and related information.

Architecture

flowchart TB
    subgraph Environment
        DotEnv[".env File\nStore API Tokens"]
    end

    subgraph Data
        PopulationCSV["population.csv\nPopulation Data"]
    end

    subgraph LLM Providers
        HuggingFace["Hugging Face\nInference API\nMistralAI/Mistral-7B-Instruct"]
        MistralAI["MistralAI\nministral-8b-latest"]
    end

    subgraph Query Engines
        PopulationQueryEngine["PandasQueryEngine\nPopulation Data"]
        DemographicsEngine["Demographics\nQuery Engine"]
        NoteEngine["Note Engine"]
    end

    subgraph Agent
        ReActAgent["ReActAgent\nMulti-Tool Reasoning"]
    end

    subgraph User Interaction
        UserInput["User Prompt Input"]
        AgentOutput["Agent Query Result"]
    end

    DotEnv --> HuggingFace
    DotEnv --> MistralAI
    
    PopulationCSV --> PopulationQueryEngine
    
    HuggingFace --> PopulationQueryEngine
    MistralAI --> ReActAgent
    
    NoteEngine --> ReActAgent
    PopulationQueryEngine --> ReActAgent
    DemographicsEngine --> ReActAgent
    
    UserInput --> ReActAgent
    ReActAgent --> AgentOutput

Loading

Features

• Multi-tool AI reasoning with ReAct Agent architecture
• Query population data from CSV sources
• Integrate multiple query engines (population, demographics)
• Support for advanced language models
• Flexible and extensible design

Technologies

• Python
• LlamaIndex
• Pandas
• Hugging Face Inference API
• MistralAI Language Models
• Dotenv for environment management

Prerequisites

• Python 3.8+
• Hugging Face Account
• MistralAI API Token
• pip package manager

🔧 Installation

1. Clone the repository:

git clone https://github.com/yourusername/AskAtlas.git
cd AskAtlas

2. Create a virtual environment:

python -m venv venv
source venv/bin/activate  # On Windows, use `venv\Scripts\activate`

3. Install dependencies:

pip install -r requirements.txt

4. Create a .env file with your tokens:

HF_TOKEN=your_hugging_face_token
MISTRAL_TOKEN=your_mistral_api_token

Usage

Run the AI agent:

python main.py

Example interactions:

Enter a prompt (q to quit): What is the population of China?
Enter a prompt (q to quit): Compare population growth between India and Brazil
Enter a prompt (q to quit): q  # to exit

Components

• Population Query Engine: Analyzes population data from CSV
• Demographics Engine: Provides demographic insights
• Note Engine: Adds contextual information
• ReAct Agent: Intelligently routes and answers queries

Configuration

Modify main.py to:

• Change language models
• Add more query engines
• Customize agent behavior

Data Sources

• data/population.csv: Primary population dataset

🤝 Contributing

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

License

This project is open source and available under the MIT License.

Author

This project is created and maintained by Ankit Mishra.

You can find my GitHub profile here: @ankitmishralive.