System Requirements Document (SRD)

Project Name: purple-ai

1. Introduction

The purple-ai project aims to develop an AI-powered electronic device designed to assist individual farmers in optimizing their agricultural practices. By leveraging advanced AI technologies, this device will analyze soil samples with minimal user input, such as geographical location and traditional crops, and provide actionable insights on crop suitability and necessary treatments.

This document outlines the system requirements for the purple-ai project, ensuring the solution is tailored to the needs of farmers in India, where agriculture plays a vital role in the economy. The device will support both online and offline modes, making it accessible even in remote areas with limited connectivity.

2. System Overview

The purple-ai device is a portable, user-friendly solution for soil testing and analysis. It integrates AI capabilities to deliver real-time recommendations for commercial crop suitability and soil treatment. Key features include:

• Dual-mode operation: Online mode for advanced cloud-based analytics and offline mode for on-device AI processing.
• Built-in screen display: Provides instant feedback on soil test results.
• Printout capability: Generates a physical summary of soil test results for farmers to keep as a reference.
• User inputs: Geographical location, traditional crops, and other relevant data.
• Outputs: Crop suitability recommendations and treatment suggestions.

The device is designed to empower individual farmers by providing actionable insights, promoting sustainable farming practices, and improving crop yields.

3. Functional Requirements

As story points:

• As a Farmer, I should be able to input my geographical location into the device.
• As a Farmer, I should be able to specify traditional crops grown in my region.
• As a Farmer, I should be able to test soil using the device and receive recommendations for commercial crop suitability.
• As a Farmer, I should be able to receive suggestions for soil treatment if required.
• As a Farmer, I should be able to view soil test results on a built-in screen.
• As a Farmer, I should be able to print a physical summary of the soil test results.
• As a Farmer, I should be able to use the device in offline mode for instant results.
• As a Farmer, I should be able to use the device in online mode for advanced analytics and updates.

4. User Personas

1. Farmer

• Description: Individual farmers who rely on actionable insights to improve crop yields and soil health.
• Goals: Optimize farming practices, reduce resource waste, and increase profitability.
• Pain Points: Limited access to advanced agricultural tools, connectivity issues in remote areas, and lack of actionable data.

5. Visuals Colors and Theme

Color Palette:

The color palette for purple-ai reflects the agricultural theme while emphasizing technology and innovation.

• Background: #F4F1E9 (Soft beige, representing soil and earth)
• Surface: #D9EAD3 (Light green, symbolizing crops and growth)
• Text: #3A3A3A (Dark gray, ensuring readability)
• Accent: #6A1B9A (Deep purple, representing innovation and the project identity)
• Muted Tones: #A8A8A8 (Neutral gray, for secondary elements)

6. Signature Design Concept

Concept: Interactive Soil Map with Real-Time Analysis

The homepage of the purple-ai device interface will feature an interactive soil map. Farmers can input their geographical location, and the map will dynamically display soil health metrics, crop suitability zones, and treatment recommendations.

Key Features:

• Dynamic Animations: The soil map will feature subtle animations, such as soil particles shifting and crops sprouting, to create a visually engaging experience.
• Micro-Interactions: Hovering over different zones on the map will reveal detailed insights about soil health and crop suitability.
• Color Shifts: The map colors will change based on soil health metrics, transitioning from red (poor health) to green (optimal health).
• Print Preview: A button on the interface will allow farmers to preview and print a summary of their soil test results.

This design concept ensures the device is not only functional but also visually captivating, making the user experience memorable and intuitive.

7. Non-Functional Requirements

• Portability: The device must be lightweight and easy to carry.
• Durability: The device should withstand harsh environmental conditions, including dust and moisture.
• Battery Life: The device must operate for at least 8 hours on a single charge.
• Connectivity: Support for Wi-Fi and Bluetooth for online mode.
• Localization: The device interface should support multiple Indian languages for accessibility.

8. Tech Stack

Frontend

• React for the web interface.

Backend

• Python with FastAPI for efficient processing.

Database

• MySQL for structured data storage.
• MongoDB for unstructured data.
• WeaviateDB for vector-based AI data.

AI Models

• GPT 5.4 for user-friendly responses.
• Claude 4.6 Opas for academic or coding work.
• Gemini 3.1 Pro for friendly responses.
• Google Nano Banana for image generation.

AI Tools

• Litellm for LLM routing.
• Langchain for AI workflows.

Orchestration

• Docker and docker-compose for local orchestration.
• Kubernetes for server-side orchestration.

9. Assumptions and Constraints

Assumptions

• Farmers have basic knowledge of operating electronic devices.
• Soil testing kits will be compatible with the device.
• Internet connectivity may be intermittent in rural areas.

Constraints

• The device must be affordable for individual farmers.
• Limited screen size for portability.
• Printout feature must use compact, low-cost printing technology.

10. Glossary

• AI: Artificial Intelligence, used for analyzing soil data and providing recommendations.
• Offline Mode: Functionality allowing the device to operate without internet connectivity.
• Online Mode: Functionality allowing the device to sync with cloud-based analytics.
• Geographical Location: The user's location, used to tailor recommendations based on regional data.
• Traditional Crops: Crops historically grown in the user's region, used as input for analysis.