Mar 2026 – Jun 2026 (Seoul)

Respondy – AI-powered Chat Assistant

An AI-powered desktop application that analyzes chat screenshots and provides context-aware reply suggestions using Google Gemini.

Watch Demo View PPT

Overview

Respondy is an AI-powered desktop chat assistant that provides real-time, context-aware reply suggestions from chat screenshots. Users can upload screenshots, receive AI-generated responses, and review conversation history within the desktop application.

The application streamlines everyday messaging by combining real-time screenshot analysis with Gemini-powered response generation in a desktop Electron environment.

Tech Stack

Electron Next.js React REST API Google Gemini

Architecture Diagram

ERD

Workflow

Users upload a chat screenshot through the desktop application.
The frontend sends the image to backend services via REST APIs.
The backend analyzes the screenshot in real time and generates reply suggestions using Gemini.
Suggested replies and conversation history are displayed in the desktop application.

Product Walkthrough

Respondy home screen — **01 · Home** Dashboard entry point for core Respondy workflows

Respondy real-time analysis screen — **02 · Real-time Analysis** Detect chat screens, analyze context, and recommend replies

Respondy manual input screen — **03 · Manual Input** Enter situation and message context for AI reply suggestions

Respondy AI chat screen — **04 · AI Chat** Practice replies with a conversational AI partner

Respondy my page screen — **05 · My Page** Manage profile, analysis history, and contact personas

Key Features

Real-time screenshot analysis
Real-time AI reply suggestions
Conversation history
Desktop application built with Electron
REST API integration

My Contributions

Developed the desktop frontend using Electron, Next.js, and React.
Implemented user interfaces for screenshot upload, AI-generated reply suggestions, and conversation history.
Integrated REST APIs to communicate with backend services for screenshot analysis and AI-generated responses.
Improved the desktop user experience by implementing responsive and intuitive interfaces.

Challenges

Building a frontend that could display AI-generated replies with minimal delay after screenshot upload.
Integrating Electron with REST APIs while maintaining a smooth desktop user experience.
Managing asynchronous image uploads and response updates efficiently.

What I Learned

Learned how to build desktop applications using Electron, Next.js, and React.
Gained practical experience integrating real-time REST API workflows into a desktop application.
Improved my understanding of asynchronous state management and API communication.
Learned how frontend applications interact with AI-powered backend services.

Demo Video

Presentation

1 / 21