Selected Project

JTPA Community Hub Project Description

Overview I built JTPA Community Hub as a bilingual community operations platform for Bay Area AI / JTPA. The platform brings together events, RSVPs, waitlists, QR check in, speaker materials, project showcases, member...

Web
Slide 1
Slide 2
Slide 3
Client
JTPA
Industry
Volunteering
Date
Jun 12, 2026
Categories
Web

Overview

I built JTPA Community Hub as a bilingual community operations platform for Bay Area AI / JTPA. The platform brings together events, RSVPs, waitlists, QR check-in, speaker materials, project showcases, member posts, Q&A, comments, likes, admin reviews, exports, and role-based management in one product.

My Role

I led the product design and full-stack implementation, including the Next.js architecture, Firebase data model, authentication flow, server-side authorization, admin tools, review workflows, and bilingual user experience.

Key Features

  • Event listings and event detail pages
  • RSVP, waitlist, and capacity tracking
  • QR-code check-in for event-day attendance
  • Speaker profiles, abstracts, slide links, and video links
  • AI project showcase submissions
  • Admin review workflows for projects and posts
  • Member posts, Q&A, comments, likes, and polls
  • Google sign-in and public profiles
  • Japanese and English localization
  • Admin exports, role management, and attendance correction

Technical Highlights

The app uses Next.js, React, TypeScript, Firebase App Hosting, Firestore, Firebase Auth, and Firebase Storage. Core application writes go through controlled server-side actions instead of direct client-side database writes, which keeps authorization, moderation, and operational state easier to reason about.

AI-Assisted Development

I used Codex and Claude Code to accelerate implementation while keeping product decisions, security boundaries, review workflows, and verification under human control. AI was most useful when given concrete constraints: existing files to follow, authorization rules to preserve, tests to pass, and clear edge cases to handle.

What Was Hard

The hardest part was balancing a simple public community experience with strict operational behavior behind the scenes. RSVP limits, waitlists, check-ins, review status, permissions, file metadata, and localization all needed to work together without exposing private or pending content.

Outcome

The result is a real deployed community platform that supports both the public face and operational backbone of Bay Area AI / JTPA.