Skip to content
BACK TO SYSTEM
02AI × E-COMMERCE × SALES

Thodar AI

An AI-powered sales assistant tailored for e-commerce, designed to understand customer intent and provide contextual product recommendations.

THE BUSINESS PROBLEM

E-commerce platforms are good at searching products, but customers don't always know what to search for.

They describe a requirement in natural language, while the system expects keywords.

That gap creates friction.

THE IDEA

Thodar AI turns product discovery into a conversation.

A customer can describe what they need, and the system converts that requirement into relevant product recommendations.

THE BUSINESS & PROCESS ANALYSIS

I mapped the buying journey: Need → Search → Compare → Decide.

The biggest friction was at search, where customers had to translate their needs into product-friendly keywords.

THE REQUIREMENTS & IMPLEMENTATION

Key requirements included intent detection, requirement extraction, product matching and conversational recommendations.

The prototype used Next.js, TypeScript, Tailwind CSS and Gemini.

THE RESULT / CURRENT STATUS

Thodar AI reached the prototype stage as an AI-powered sales assistant.

The project showed how AI can reduce the gap between what customers mean and what systems expect.

WHAT I LEARNED

The best AI experiences don't make users adapt to the software.

The software adapts to the user.

TECHNOLOGY

Product conceptAI interactionWorkflowTechnical implementation