price - bot

Introduction

The "price - bot" project aims to develop a modern AI-powered website that provides users with the ability to upload assignment files related to various courses (bachelor, diploma, etc.) and receive a price estimate for the assignment. This tool is designed to cater to academic institutions, providing a streamlined process for evaluating assignment costs.

System Overview

The "price - bot" system is an AI-driven platform that allows users to upload academic assignments and receive pricing information. The system leverages advanced AI models to analyze the content and complexity of the assignments to provide accurate pricing. The platform is intended for use by students and academic institutions, offering a user-friendly interface and efficient processing capabilities.

Functional Requirements

• As a User, I should be able to upload an assignment file to the platform.
• As a User, I should receive a price estimate for the uploaded assignment file.
• As a User, I should be able to specify the type of course (bachelor, diploma, etc.) related to the assignment.
• As a User, I should receive feedback on the complexity and content of the assignment.
• As a User, I should be able to view a summary of the pricing criteria used for the estimate.

User Personas

• Student: A user who uploads assignments to receive pricing estimates.
• Academic Institution Staff: A user who manages and oversees the pricing process for assignments.

Core User Flows

• Student uploads assignment file -> selects course type -> receives price estimate -> reviews pricing criteria summary.
• Academic Institution Staff reviews uploaded assignments -> verifies pricing estimates -> provides feedback to students.

Visuals Colors and Theme

• primary: #2A9D8F (Teal)
• primary_light: #A8DADC (Light Teal)
• secondary: #E76F51 (Coral)
• accent: #F4A261 (Orange)
• highlight: #E9C46A (Gold)
• bg: #F1FAEE (Off White)
• surface: rgba(42, 157, 143, 0.8)
• text: #264653 (Dark Slate)
• text_muted: #8D99AE (Muted Gray)
• border: rgba(38, 70, 83, 0.2)

Signature Design Concept

The homepage of the "price - bot" will feature an interactive "Assignment Carousel." This carousel will allow users to drag and drop assignment files directly onto a rotating 3D model of a book. As users interact with the carousel, the book will open to reveal a dynamic pricing dashboard. This dashboard will animate in real-time, showing the breakdown of pricing criteria. The use of @react-three/fiber and @react-three/drei will enable the 3D interactions, while framer-motion will handle the smooth transitions and animations.

Interaction Model & Motion Direction

The landing page will utilize a "parallax" interaction model. Layers of educational-themed graphics will move at different speeds as users scroll, creating a sense of depth and engagement. Key sections will feature scroll-triggered animations to reveal content, ensuring a visually rich and interactive first impression.

Non-Functional Requirements

• The system must process assignment uploads and provide pricing estimates within 30 seconds.
• The platform should support file uploads up to 50MB in size.
• The website must be accessible and responsive across all modern browsers and devices.

Tech Stack

• Frontend: React for Web
• Backend: Python, FastAPI
• Database: MySQL or MariaDB
• AI Models: Claude Sonnet 5 for academic or coding work
• AI Tools: Litellm for LLM Routing, Langchain
• Local Orchestration: Docker, docker-compose
• Server-side Orchestration: Kubernetes

Assumptions and Constraints

• The system assumes that users have valid credentials to access the platform.
• The platform is constrained to operate within the academic institution's network for security reasons.
• Pricing estimates are based on predefined criteria and may not reflect actual market rates.

Glossary

• AI: Artificial Intelligence
• LLM: Large Language Model
• 3D Model: Three-dimensional representation used for interactive elements
• Parallax: A scrolling technique that creates an illusion of depth by moving background images slower than foreground images.