Public Client Work

Playjump

Worked on an AI video creation platform focused on helping creators and teams generate, style, and produce videos with AI.

Built under AI Creators Studio.

Last updated: May 17, 2026.

01

Context

Playjump was an AI video creation platform for creators, marketers, and production teams.

I worked on the product for around 7-8 months through AI Creators Studio.

Because the company is no longer active, the live product is unavailable, but public references to the company and product still exist.

02

My Role

Contributed as a product-focused frontend engineer through AI Creators Studio.

Worked across product UI, creator workflows, generation surfaces, routing, state handling, reliability, and production polish.

Collaborated inside an existing team and codebase rather than presenting Playjump as my own product.

03

What I Worked On

AI creation flows for video, image, character, and style-related product surfaces.

Creator-facing galleries, preview modals, empty states, navigation, layout improvements, and saved-output views.

Credit checks, guarded generation actions, moderation/admin surfaces, support tooling, error monitoring, and analytics instrumentation.

04

Technical Scope

9 tools

Next.js App Router frontend with React, Tailwind CSS, Radix UI, TanStack Query, Redux, and form-driven product surfaces.

Firebase-backed authentication state, protected user workflows, credit-aware actions, and asynchronous AI generation states.

Upload and media flows connected to Cloudflare R2-style storage, API routes, moderation review paths, Sentry, PostHog, and customer-support tooling.

  • Next.js
  • React
  • Tailwind CSS
  • TanStack Query
  • Redux
  • Firebase
  • Sentry
  • PostHog
  • Cloudflare R2 uploads
05

Challenges

Keeping AI generation flows understandable while outputs could be slow, fail, need review, or require enough user credits.

Maintaining a fast creator UI across galleries, previews, route-level modals, media-heavy screens, and repeated product iterations.

Shipping inside an active product team while preserving private client code, internal data, and unreleased product details.

06

What I Learned

AI media products need strong state design around pending, failed, reviewed, saved, and credit-gated outputs.

Creator tools work better when the product makes generated assets easy to find, preview, reuse, and understand.

A public case study can still be useful without exposing private implementation details when the scope, role, constraints, and outcomes are clear.

07

Public References

08

Confidentiality Note

Parts of the engagement are private, and the company is no longer active. This case study avoids private code, internal data, unreleased product details, user data, and business-sensitive implementation details.