WellyGeo - AI Brand Visibility Tracking Platform

11/2025 ~ Present

Frontend Development
Crawler System
AI
SaaS
Data Visualization
Active
Featured

Track brand visibility performance in AI search engines such as ChatGPT, Google Gemini, and Google AI Overview, providing competitor analysis and trend reports

WellyGeo - AI Brand Visibility Tracking Platform

Tech Stack

React 19

Frontend framework with Vite for building SPA data analytics dashboard

TypeScript

Unified type system for frontend and backend ensuring code quality

shadcn/ui + Tailwind CSS

Frontend component library and styling framework for building data visualization interfaces

Recharts

Chart visualization for presenting brand visibility trends and analytics

i18next

Internationalization framework supporting Traditional Chinese and English

Playwright

Browser automation engine for crawling ChatGPT and Gemini AI responses

Express.js

Crawler backend API server handling crawler task queues

SerpAPI

Google AI Overview search result data source

Docker

Containerized deployment integrating Playwright browser environment

AWS Batch / Fargate

Serverless container execution environment for scalable crawler tasks

AWS S3

Cloud storage for crawled results (JSONL format)

AWS CDK

Infrastructure as Code for managing cloud infrastructure

Project Overview

WellyGeo is an AI Brand Visibility Tracking Platform (GEO, Generative Engine Optimization) developed by Solwen AI, helping brands monitor their exposure performance in AI search engines.

The system automatically sends queries to ChatGPT, Google Gemini, and Google AI Overview, captures AI response content and analyzes how frequently brands are mentioned, citation sources, and competitor comparisons. It produces visual trend reports to help marketing teams optimize their brand visibility strategy within the AI ecosystem.

System Architecture

WellyGeo consists of two major subsystems forming a complete data collection and analysis pipeline:

Frontend — Data Analytics Dashboard

An SPA application built with React 19 + TypeScript + Vite, integrated with the shadcn/ui component library and Recharts chart visualization. Includes feature modules for project management, brand visibility trend charts, AI response analysis, citation source tracking, and competitor comparison, with i18next bilingual support for Traditional Chinese and English.

Crawler — AI Response Crawling Engine

A crawler service built with Playwright + Express.js that simulates user interactions through browser automation to capture responses from ChatGPT and Google Gemini. Google AI Overview data is obtained via SerpAPI. Crawled results are uploaded to AWS S3 in JSONL format, with support for large-scale distributed execution on AWS Batch/Fargate and Webhook callback notifications.

Core Features

Brand Visibility Analysis

Tracks trends in how frequently a brand is mentioned across different AI platform responses, presenting visibility distribution through line charts, pie charts, and bar charts, along with detailed data tables.

AI Response Content Analysis

Analyzes the context in which a brand appears in AI responses, sentiment orientation, and recommendation ranking, revealing how AI describes and recommends the brand.

Citation Source Tracking

Analyzes the web sources and type distribution cited in AI responses, identifying which content AI uses as the basis for brand information.

Competitor Comparison

Automatically detects or manually configures competitors to compare brand and competitor AI visibility differences across the same query dimensions.

AI Keyword Generation

Integrates AI to automatically generate brand-related search topics and query phrases, with CSV batch import support to accelerate project setup.

Multi-Platform Crawler Scheduling

Supports three major platforms — ChatGPT, Google Gemini, and Google AI Overview — with flexible platform combination selection via the AI_PLATFORMS parameter.

My Role & Contributions

Full Frontend Development

Independently responsible for the complete development of the Frontend dashboard, including project management workflows, brand visibility charts (line, pie, and bar charts), AI response analysis pages, citation source tracking, competitor comparison configuration, and i18next bilingual switching between Traditional Chinese and English.

Crawler System Development Support

Participated in Crawler backend development, helping build Playwright browser automation crawlers, fixing Gemini error handling, and maintaining documentation and logging systems.

Development Workflow Setup

Introduced Husky + lint-staged for automated pre-commit checks, configured ESLint + Prettier to enforce consistent code style, and established team development standards.

Code Review

Responsible for reviewing team members’ Pull Requests to ensure code quality and architectural consistency.

Technical Highlights

Anti-Detection Browser Automation

The Crawler integrates Puppeteer-Extra Stealth Plugin with custom browser fingerprinting to bypass bot detection mechanisms on AI platforms, ensuring stable crawling.

ChatGPT 2FA Automated Login

Implements TOTP (Time-based One-Time Password) automatic generation to support fully automated login with ChatGPT two-factor authentication.

AWS CDK Infrastructure as Code

Uses AWS CDK (TypeScript) to define the complete cloud infrastructure, including Batch compute environments, Fargate task definitions, IAM roles, and ECR container image repositories.

CI/CD Automated Deployment

The Frontend is automatically deployed via AWS Amplify; the Crawler uses GitHub Actions to automatically build Docker images and push to ECR on Git Tag events, triggering AWS Batch updates.