Launchpad AI Review System (L-AIRS)

Project Overview

Industry: Technology/Education Developer: Launchpad Cohort 1 (Lead: [Kristian]) Completion Date: 06/27/2025 GitHub Repository: [https://github.com/DeniedPath/Launchpad-AI-Review-System.git] Trello: [https://trello.com/invite/b/682cbc33d8864d5cfd543fad/ATTIf793605f1c1ae34e0dff0769d9f40f2d9F8AA788/launchpad-ai-review-system] Live Demo: [launchpad-ai-review-system.vercel.app]

---

Business Problem

Problem Statement

Launchpad students need timely, structured feedback on their Python code to accelerate learning and improve coding skills. Manual review by instructors is time-consuming and inconsistent, leading to delays and uneven feedback quality. There is a need for an automated, scalable solution that provides actionable, rubric-based feedback without revealing full solutions.

Target Users

• Students: Coding bootcamp participants with beginner to intermediate Python skills, seeking instant, actionable feedback.
• Admins/Instructors: Staff who create and manage assignments, monitor student progress, and ensure feedback quality.

Current Solutions and Limitations

• Manual code review by instructors (slow, inconsistent)
• Limited or generic feedback from static linters or auto-graders
• No unified dashboard for assignment management and feedback delivery

---

Solution Overview

Project Description

L-AIRS is a web-based platform that automates code review for Python assignments using AI. Students submit code via a dashboard, select assignments, and receive instant, rubric-based feedback powered by PlayLab AI. Admins can create, edit, and manage assignments through a protected admin panel. The system ensures feedback is structured, actionable, and never reveals full solutions, supporting scalable, high-quality learning.

Key Features

• Student dashboard for code submission and feedback
• Assignment selection with requirements display
• Markdown-formatted, rubric-based AI feedback
• Admin CRUD for assignments (protected)
• Role-based authentication (admin/student)

Value Proposition

L-AIRS delivers instant, high-quality feedback at scale, reducing instructor workload and improving student outcomes. Its rubric-based approach ensures consistency, and the modern UI/UX makes it accessible and engaging.

AI Implementation

• Uses PlayLab API to analyze student code and generate structured feedback
• AI provides rubric-based evaluation (loops, conditionals, etc.)
• No code solutions are revealed, only actionable feedback

Technology Stack

• Frontend: Next.js, React, Radix UI
• Styling: Tailwind CSS
• Backend: Next.js API Routes, Prisma ORM
• Database: PostgreSQL (Neon.tech)
• Authentication: NextAuth.js (credentials, Prisma adapter)
• AI Services: PlayLab API
• Deployment: Vercel/Netlify (recommended)
• Other Tools: Prisma, ReactMarkdown, ESLint, Prettier

---

Technical Implementation

Wireframes & System Architecture

[Student] <-> [Next.js Frontend] <-> [API Routes] <-> [Prisma ORM] <-> [PostgreSQL]
                                            |
                                            v
                                    [PlayLab AI API]
[Admin]   <-> [Admin Panel (protected)]

• Students interact with the dashboard to submit code and view feedback.
• Admins access protected routes to manage assignments.
• All data flows through API routes, with Prisma handling DB access and PlayLab providing AI feedback.

Database Schema

See DATABASE_SCHEMA_DIAGRAM.md for a full diagram.

• AdminUser (id, email, passwordHash, createdAt)
• Assignment (id, title, requirements, description, rubric, createdAt, updatedAt, creatorId)
• Submission (id, assignmentId, code, feedback, rating, createdAt)

AI Model Details

• Model(s) Used: PlayLab API (cloud-based, proprietary)
• Purpose: Analyze student code, provide rubric-based feedback
• Integration Method: API calls from backend
• Model Performance Metrics: N/A (external API)

Key Components and Code Snippets

• StudentDashboard: Handles assignment selection, code/file upload, and feedback display
• AdminEditAssignment: Protected form for editing assignments
• AI Integration:

const res = await fetch("/api/ai", { method: "POST", body: JSON.stringify({ code, assignment }) });
const data = await res.json();
setFeedback(data.feedback);

• Authentication:

import { getServerSession } from "next-auth";
// ...
const session = await getServerSession(authOptions);
if (!session || !session.user.isAdmin) redirect("/pages/login");

• API Routes: | Endpoint | Method | Purpose | Auth Required | |-------------------------|--------|----------------------------|---------------| | /api/assignments | GET | List assignments | No | | /api/assignments | POST | Create assignment | Yes (admin) | | /api/assignments?id=... | PUT | Update assignment | Yes (admin) | | /api/assignments?id=... | DELETE | Delete assignment | Yes (admin) | | /api/ai | POST | Get AI feedback | No |

---

User Interface and Experience

User Journey

1. User arrives at dashboard
2. Selects assignment, submits code or file
3. Receives instant AI feedback
4. Admins log in to manage assignments

Key Screens and Components

• Dashboard: Assignment selection, code input, feedback display
• Admin Panel: Assignment management (create, edit, delete)
• Login: Admin authentication

Responsive Design Approach

• Uses Tailwind CSS and Radix UI for mobile-friendly, accessible layouts

Accessibility Considerations

• Semantic HTML, keyboard navigation, color contrast, ARIA labels

---

Testing and Quality Assurance

• Manual and exploratory testing
• Linting and formatting enforced (ESLint, Prettier)
• Known Issues: No automated tests yet; future work planned

---

Deployment

Deployment Architecture

• Hosted on Vercel/Netlify (recommended)
• Environment variables required:
  • DATABASE_URL
  • NEXTAUTH_SECRET
  • NEXTAUTH_URL
  • NEXT_PUBLIC_PLAYLAB_API_KEY
  • NEXT_PUBLIC_PLAYLAB_PROJECT_ID

Build and Deployment Process

1. Set environment variables
2. Run npm run build
3. Deploy to Vercel/Netlify

---

Future Enhancements

• Add automated and integration tests
• Student authentication and progress tracking
• Enhanced admin analytics
• More granular AI feedback and rubric customization
• Scalability improvements for larger cohorts

---

Lessons Learned

• Integrating AI feedback requires careful prompt engineering
• Prisma and NextAuth.js simplify full-stack development
• UI/UX polish is critical for adoption
• Automated testing and CI/CD should be prioritized early

---

Project Management

• Development tracked in Trello
• Used GitHub for version control
• Resources: Prisma docs, Next.js docs, PlayLab API docs

---

Conclusion

L-AIRS delivers scalable, high-quality code review for Launchpad students, reducing instructor workload and improving learning outcomes. The project demonstrates the power of modern full-stack frameworks and AI integration in education.

---

Appendix

Setup Instructions

# Clone the repository
git clone [repository URL]
cd launchpad-code-review-tool

# Install dependencies
npm install

# Set up environment variables
cp .env.example .env.local
# Edit .env.local with your values

# Run development server
npm run dev

Additional Resources

• [PlayLab API Documentation]
• [Next.js Documentation]
• [Prisma Documentation]