The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
A lot of AI image/video tools are expensive.
That’s not what bothers me.
What bothers me is when a tool charges premium pricing while still missing the workflow depth that would justify it.
That’s my issue with ElevenLabs Flows.
https://www.scopeful.org/blog/elevenlabs-pricing-review
scopeful.org
Half the Invoice: Why ElevenLabs is a $300 Pricing Trap
How ElevenLabs prices AI image and video generation: a review of Flows, and the credit cliff steep enough to triple your bill between one image and the next.
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Related posts
check out how this AI lead enrichment + qualification workflow does its job!
Nice build, and good call putting a Structured Output Parser after Gemini. How do you keep the score itself stable from one run to the next? I'd split it into the reasons (company size, role, budget signal) as separate fields and compute the final number in a Code node, so the...
Advizi AI — AI Thumbnail & Video Generator SaaS
Project Overview
Advizi AI is a personal SaaS project I built to help content creators generate visual content for YouTube and other platforms using AI. The platform brings thumbnail and video generation into one application, with user accounts, a credit-based usage system, and subscription billing.
The idea was to build more than a simple AI generation interface. I wanted to create a full-stack application where users could access generation features, use credits, and choose a plan that fits their needs.
What I Built
AI Thumbnail Generation
Implemented AI-powered thumbnail generation for YouTube and other platforms, allowing users to create visual content and save or download their generated results.
AI Video Generation
Built working video generation functionality alongside thumbnail creation, bringing two content generation workflows into one platform.
User Authentication
Integrated Clerk to handle user authentication and access to the application.
Credit-Based Usage System
Implemented a credit system with three plans:
Free: 20 credits
Pro: 80 credits
Premium: 240 credits
Subscription Billing
Integrated Stripe for subscription and payment workflows connected to the platform's plans.
Media Handling and Error Monitoring
Integrated Cloudinary for media handling and Sentry for error monitoring.
My Role:
I developed Advizi AI as a personal full-stack project, working across the frontend, backend, database, authentication, AI integration, and billing functionality.
The project involved bringing multiple services together into a single application, with React and TypeScript on the frontend and Node.js with Express.js on the backend.
Technology Stack
Frontend: React, TypeScript, Vite
Backend: Node.js, Express.js
Database: MongoDB
AI integration: Gemini API
Authentication: Clerk
Payments: Stripe
Media handling: Cloudinary
Error monitoring: Sentry
The Result
Advizi AI is a working AI content-generation application that combines thumbnail and video generation with authentication, usage credits, and subscription billing. It demonstrates my ability to develop a full-stack SaaS product and integrate AI services with the supporting systems needed for a user-facing application.
I had the pleasure of working with 21st.dev to make small design fixes easy to see.
AI can build a page, but the details don’t always match. Different button sizes. Inconsistent text. Selected states that aren’t clear. The design feels off.
To explain Design Bug Bot, I built a UI component with those problems and animated the fixes one at a time.
I brought each issue into focus, showed what changed, then pulled back to show the result.
Why this approach?
People won’t pause and replay your video to figure out what changed.
That’s why I make each feature obvious, one at a time.
If a feature still needs explaining after you’ve shown it, the video hasn’t done its job.
