AI-Powered Multilingual WhatsApp Tutor

Title:
Build AI WhatsApp Tutor (ChatGPT + WhatsApp Business API | Multilingual | Education Bot | JSON-based Flow)

Summary:

We are creating a WhatsApp-based AI Tutor System that answers students’ academic and government-exam questions using ChatGPT in the language they are most comfortable with.

This is not a random chatbot.
It’s a structured AI-powered education assistant with class-wise, subject-wise, and topic-wise organization (based on NCERT and government exam patterns).

We already have the Master Prompt that generates the entire dataset (class → subject → chapter → topic → response options in English, Hindi, and Hinglish).
You’ll convert that structured data and logic into a functional, API-connected WhatsApp Tutor integrated with OpenAI and WhatsApp Business API.

What We’re Building

A WhatsApp Tutor that:

Greets users and collects:

Class (6–12 or Government Exam)

Subject (Science, Maths, SST, etc.)

Preferred Language (English, Hindi, Hinglish)

Waits for the user’s question (typed or image).

Passes it to ChatGPT with contextual metadata (class, subject, and language).

Returns a clean, age-appropriate answer formatted for WhatsApp readability.

Suggests next-step options after every answer:

Explain with Example

Give Me a Short Quiz

Next Topic

Back to Main Menu

Functional Expectations

Platform: WhatsApp Business API (Twilio, Gupshup, or 360Dialog)
Backend: Node.js or Python
AI Layer: OpenAI GPT-4 or GPT-3.5 API
Storage: Google Sheets or Airtable (to log class, subject, language, question, timestamp)
Dataset: JSON format (auto-generated from Master Prompt)
Response Handling: Multi-turn, session-aware (remembers user language and last topic)
Languages Supported: English, Hindi, Hinglish

Example Flow

User joins →
"Welcome to your personal AI tutor!"
"Choose your category:"

Class 6–12

Government Exam

User: 8
Bot: "Choose subject" → Science
Bot: "Choose your language" → Hindi
User: "Friction kya hota hai?"
Bot: "Friction यानी घर्षण वो बल है जो किसी वस्तु की गति का विरोध करता है। जैसे कि अगर आप किताब को मेज पर धक्का देते हैं, तो मेज की सतह का घर्षण बल उसे रोकने की कोशिश करता है।"
Options: Explain with Example | Quiz | Next Topic | Menu

Deliverables

Fully functional WhatsApp bot connected to ChatGPT (GPT-4 or GPT-3.5)

End-to-end flow (Class → Subject → Language → Q&A → Options)

JSON dataset integration from the provided Master Prompt

Data logging via Google Sheets or Airtable

Multilingual support (English, Hindi, Hinglish)

Demo video and documentation (setup, code, API keys, flow explanation)

Tech Stack Preference

Node.js (Express) or Python (Flask/FastAPI)
OpenAI API integration
WhatsApp Business API (Twilio, Gupshup, or 360Dialog)
Google Sheets API or Airtable API
JSON / NoSQL structure for topic mapping
Optional: Pabbly, Make, or Zapier for automation

Timeline & Workflow

Phase 1 – API setup and WhatsApp connection (3 days)
Phase 2 – ChatGPT and language integration (4 days)
Phase 3 – Class/subject/topic flow setup (3 days)
Phase 4 – Testing and documentation (2 days)
Total: Fully functional MVP in 10–12 days

Budget
$50 fixed for the MVP version.
Bonus available for early delivery and clean, well-documented code.

Ideal Developer

Experience building WhatsApp chatbots (share demo or screenshots)

Skilled with multi-language prompt handling and dynamic JSON responses

Familiar with OpenAI integration best practices (token efficiency, latency, and cost control)

Can design scalable, modular bot architecture

Provides daily progress updates and clear communication

Test Task (For Shortlisting)

Create a small WhatsApp flow or mock API response:
Class: 8
Subject: Science
Language: Hindi
Question: "Friction kya hota hai?"
Response should explain friction clearly and display options (Explain, Quiz, Next Topic, Menu).

This test helps demonstrate understanding of multilingual flow and formatting.

Apply With

Relevant project links or demos

Preferred tech stack

Timeline estimate

Any clarifying questions