Skip to content

Qesem is an AI-powered study assistant. Students can upload PDFs or text notes, chat with an agentic assistant, generate quizzes automatically, and receive graded results.

License

Notifications You must be signed in to change notification settings

CSEC-ASTU/Qesem

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

78 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Qesem — AI Study Assistant

Tech Stack Tech Stack Tech Stack Tech Stack Tech Stack AI SSE License Build

Table of Contents

Overview

Qesem is an AI-powered study assistant. Students can upload PDFs or text notes, chat with an agentic assistant, generate quizzes automatically, and receive graded results. Responses stream in real time via Server-Sent Events (SSE), and answers are grounded with Retrieval-Augmented Generation (RAG) over your uploaded materials.

Features

  • RAG over uploads: Chunking and vector embeddings (Voyage AI) stored in MongoDB for grounded responses.
  • Agentic flows: LangChain + LangGraph for quiz generation, grading, and contextual reasoning.
  • Realtime SSE: Streaming assistant responses for smooth UX.
  • Quiz automation: Auto-generate quizzes and grade user-submitted answers.
  • File uploads: PDF/TXT ingestion with size limits and MIME validation.
  • Session history: Persisted conversations with resume capability.
  • Configurable LLMs: Gemini model selection via env; Voyage embed model override.
  • Dark UI: React + Vite + Tailwind CSS 4 with responsive layout.

Tech Stack

  • Frontend: React 19, TypeScript, Vite, Tailwind CSS 4, Zustand.
  • Backend: Node.js, Express 5, TypeScript, MongoDB (Mongoose), SSE.
  • AI/ML: LangChain, LangGraph, Gemini (Google Generative Language API), Voyage embeddings.
  • Build/Tooling: Vite, ESLint, ts-node, TypeScript.

Architecture

  • client/: Vite React SPA; chat UI, quiz UI, SSE client, upload trigger.
  • server/: Express API; routes for chat, quiz, upload, sessions, SSE; LLM + embedding services; MongoDB persistence.

Getting Started

Prerequisites

  • Node.js 18+ (recommended)
  • npm 9+
  • MongoDB Atlas or local MongoDB
  • API keys: Google Generative Language (Gemini) and Voyage AI

Clone & Install

git clone https://github.com/CSEC-ASTU/Qesem.git
cd Qesem
npm install --prefix server
npm install --prefix client

About

Qesem is an AI-powered study assistant. Students can upload PDFs or text notes, chat with an agentic assistant, generate quizzes automatically, and receive graded results.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •  

Languages