~/asadsiddiqui

Complete

Quest AI

Quest AI is an intelligent learning companion that generates personalized roadmaps, quizzes, flashcards, and enables AI-powered interaction with study materials.

Screenshot 0
Screenshot 1
Screenshot 2
Screenshot 3
Screenshot 4
Screenshot 5
Screenshot 6
Screenshot 7
Screenshot 8
Screenshot 9
Screenshot 10
Screenshot 11
Screenshot 12
Screenshot 13

Tech Stack

Next.jsTypeScriptReactTailwind CSSPostgreSQLPrismaGemini 2.0 FlashVercel AI SDKLangChainVector DatabaseRAGEmbeddingsShadCN UI

The Problem

Traditional learning platforms lack personalization and fail to adapt to individual learning styles, pace, and goals.

  • 1.

    No personalized learning roadmap

  • 2.

    Low engagement with static study material

  • 3.

    Difficulty tracking learning progress

  • 4.

    Manual quiz and revision preparation

The Solution

Quest AI delivers an adaptive learning experience by combining AI-generated roadmaps, assessments, and contextual learning assistance.

  • 1.

    AI-generated personalized learning roadmaps

  • 2.

    Quiz and flashcard generation from uploaded content

  • 3.

    Chat with PDF using Retrieval-Augmented Generation (RAG)

  • 4.

    AI-powered learning assistant for instant explanations

  • 5.

    Progress tracking and structured learning flow

Architecture

A full-stack AI-first learning platform built on Next.js with integrated AI pipelines and vector-based retrieval.

  • 1.

    Next.js App Router for UI and server logic

  • 2.

    AI pipelines using Gemini and Vercel AI SDK

  • 3.

    Vector database for document retrieval (RAG)

  • 4.

    PostgreSQL for user data and learning progress

  • 5.

    LangChain for document processing and context retrieval

Key Technical Decisions

  • 1.

    Used Next.js as a full-stack framework

    Unified frontend and backend with excellent DX and performance

  • 2.

    Implemented RAG with a vector database

    Enabled accurate, context-aware AI responses from user documents

  • 3.

    Adopted structured AI outputs

    Ensured consistent storage and UI rendering of AI-generated content

Challenges & Resolutions

  • 1.

    Maintaining relevance in AI-generated responses

    Used embedding-based similarity search with contextual prompts

  • 2.

    Handling diverse document formats

    Standardized document parsing and chunking pipelines

  • Future Enhancements

    • 1.

      Agentic AI for adaptive learning paths

    • 2.

      Automated revision schedules

    • 3.

      Collaborative learning features

    • 4.

      Multi-language content support

    © 2026 Asad Siddiqui. All rights reserved.

    Building clean, production-ready applications.