Skip to content

Latest commit

 

History

History
392 lines (238 loc) · 4.58 KB

File metadata and controls

392 lines (238 loc) · 4.58 KB

Quantara AI

Overview

Quantara AI is a web platform that automates the creation of engineering laboratory reports in realistic handwritten notebook format.

Students upload their experiment details, readings, and handwriting samples. The platform generates a complete lab report, renders it in the student's handwriting style using blue ink, draws hand-sketched diagrams and graphs, and exports the result as a notebook-style PDF.

The goal is to eliminate the manual effort of writing lengthy lab reports while preserving the handwritten format required by many institutions.


Core Features

Student Authentication

  • Register/Login
  • Dashboard
  • Slot-based submission system
  • Request history
  • Download generated reports

Lab Report Generation

Students provide:

  • Course code
  • Experiment title
  • Aim
  • Apparatus
  • Theory notes (optional)
  • Raw observations/readings
  • Lecturer instructions
  • Preferred report format

AI generates:

  • Title page
  • Aim
  • Apparatus
  • Theory
  • Procedure
  • Observation tables
  • Calculations
  • Results
  • Discussion
  • Conclusion
  • References

Handwriting Learning

Students upload:

  • 5–10 pages of handwritten notes

System extracts:

  • Character shapes
  • Letter spacing
  • Writing slant
  • Writing size

System creates a reusable handwriting profile.


Handwritten Report Rendering

Generated report is converted into:

  • Blue pen handwriting
  • Notebook paper pages
  • Natural spacing variations
  • Line-by-line writing
  • Multi-page report format

Hand-Drawn Sketch Generation

Automatically generates:

  • Circuit diagrams
  • Block diagrams
  • Engineering apparatus sketches
  • Graphs
  • Tables
  • Flowcharts

Rendered to look hand drawn rather than computer generated.


PDF Export

Output:

  • Multi-page notebook PDF
  • Blue ink handwriting
  • Hand-drawn diagrams
  • Ready for printing or copying

Architecture

Frontend (React + Vite)

Express API

PostgreSQL

BullMQ + Redis Queue

Python AI Service (FastAPI)

Report Generator + Handwriting Engine + Diagram Renderer


Tech Stack

Frontend

  • React
  • Vite
  • TypeScript
  • TailwindCSS
  • TanStack Query
  • React Router
  • shadcn/ui

Backend

  • Node.js
  • Express.js
  • Prisma ORM
  • PostgreSQL
  • JWT Authentication
  • Multer
  • BullMQ
  • Redis

AI Service

  • Python
  • FastAPI
  • PyTorch
  • Pillow
  • OpenCV
  • NumPy

Open Source / Free Tools

Report Generation

Primary

OpenRouter

Benefits:

  • Free models available
  • OpenAI-compatible API
  • Can switch models without changing code

Recommended Models:

  • DeepSeek
  • Qwen
  • Gemma
  • Llama

Expected Cost:

  • Free during development
  • Under ₦5,000/month for early usage if using inexpensive models

Handwriting Generation

Phase 1 (MVP)

Font-based approach

Tools:

  • FontForge
  • Calligraphr
  • TraceFont

Benefits:

  • Free or low cost
  • Fast implementation
  • Good enough for MVP

Phase 2 (Advanced)

AI handwriting synthesis

Research Projects:

  • GANWriting
  • DeepWriting
  • Diffusion-based handwriting models

Benefits:

  • More realistic output
  • Can imitate individual writing styles

Drawback:

  • Higher complexity

Diagram Generation

RoughJS

Purpose:

Creates diagrams that appear hand drawn.

Used for:

  • Graphs
  • Circuits
  • Flowcharts
  • Engineering sketches

Cost:

Free and open source


Notebook Rendering

Tools:

  • Pillow
  • OpenCV

Responsibilities:

  • Blue pen effect
  • Paper texture
  • Notebook lines
  • Ink variations
  • Page composition

Cost:

Free


PDF Generation

Options:

  • pdf-lib
  • PDFKit

Cost:

Free


Job Processing

Queue System

Tools:

  • Redis
  • BullMQ

Workflow:

Student submits report

Queue job

Generate report

Generate handwriting

Generate diagrams

Generate notebook pages

Generate PDF

Notify student


Database Tables

users

handwriting_profiles

lab_reports

generated_pages

slots

payments

notifications


Estimated Monthly Cost

Development Stage:

₦0 – ₦5,000/month

Production (First 100–300 Users):

₦5,000 – ₦20,000/month

Major expenses:

  • Hosting
  • Database
  • AI API usage

Most rendering, handwriting generation, PDF creation, and diagram generation can remain completely open source.


MVP Goal

A student uploads experiment details and handwriting samples.

The system automatically:

  1. Generates a complete lab report.
  2. Renders it in blue handwritten style.
  3. Generates hand-drawn diagrams and graphs.
  4. Places everything on notebook paper.
  5. Exports a multi-page PDF.

Result:

A realistic handwritten engineering lab report generated with minimal human effort.