EduPilot

Core MVP Engine for AI-Driven Curriculum Analysis & Lesson Generation

Overview

EduPilot processes raw, unstructured curriculum PDFs and automatically transforms them into strictly formatted 60-minute lesson plans. Built for scale, it isolates data parsing, LLM inference, and UI rendering to ensure a distraction-free environment for educators.

Tech Stack

Architecture & Features

• Data Ingestion: Utilizes pdf2json to accurately map byte-streams of government-issued syllabi into structured text blocks.
• LLM Pipeline: Integrates Gemini 2.5 Flash via Next.js server endpoints to generate stateful, markdown-formatted lesson plans based on specific dynamic topics.
• Authentication & Persistence: Leverages Supabase PostgreSQL. Database operations are protected by strict Row Level Security (RLS) policies tied directly to Next.js server-side HTTP cookies.
• Routing: Implements the Next.js App Router for dynamic segment isolation (/curricula/[id], /lessons/[topic_id]).

Local Setup

Ensure Node.js 20+ is installed on your local environment.

# Clone the repository
git clone https://github.com/jezreal-dev/edupilot.git

# Navigate into the project directory
cd edupilot

# Install NPM dependencies
npm install

Environment Configuration

Create a .env.local file at the repository root and map your keys:

NEXT_PUBLIC_SUPABASE_URL=your_supabase_url
NEXT_PUBLIC_SUPABASE_ANON_KEY=your_anon_key
GEMINI_API_KEY=your_google_ai_key

Development Server

npm run dev

Security & Vulnerabilities

Please review the SECURITY.md guidelines prior to reporting any vulnerabilities found in this system.

License

Copyright (c) 2026 EduPilot. All Rights Reserved.

This codebase is strictly proprietary. Unauthorized copying, modification, or distribution is legally prohibited. See the LICENSE file for complete terms.