Creative & campaigns
Generate image and video ads, build Meta campaigns from product links, and plan social content.
PROJECT 03 / ECOMMERCE & AUTOMATION
Zyberon AI.
Engineering case study by Priyansh Dobariya · Next.js · Supabase · Shopify · n8n · Gemini · Meta APIs · Stripe
One platform for store operations.
A Shopify-focused SaaS platform connecting customer support, product pages, ad creative, campaigns, social publishing, and profit analysis through AI and automation.
AI × COMMERCEZYBERON.AI
Connected intelligence for Shopify.
Product context→AI creative→Campaigns
THE PRODUCT
One workspace. A connected AI workforce.
Zyberon brings Shopify workflows together across marketing, customer service, and business operations.
Store experiences
Create pages, quiz funnels, and lifecycle emails, with customer support informed by Shopify order context.
Business intelligence
Research products, track profit, and coordinate tasks across the store’s AI tools.
MY CONTRIBUTIONS
Behind the build.
Built the full-stack Zyberon platform from scratch.
Connected the dashboard and server actions to Shopify, Meta, Stripe, Supabase, and n8n workflows.
Implemented AI-assisted commerce modules with review, status, and integration boundaries.
Tools & technologies
MY CONTRIBUTION
What I built
I built Zyberon AI from scratch as a full-stack Shopify automation platform: the dashboard, API and server-action layer, data and authentication integration, AI workflows, and external-service connections. The platform brings several store-operations modules into one product.
THE PROBLEM
What this solves
Shopify sellers manage support messages, product pages, creatives, advertising, social posts, and profit data across separate systems. Repeating product and customer context across those tools makes each workflow slower and harder to review.
ARCHITECTURE
How it works
A Next.js dashboard and server-action layer sit over Supabase authentication, PostgreSQL data, and asset storage. Application routes and webhook utilities connect Shopify orders and products, Stripe subscription events, email integrations, and Meta APIs. n8n receives automation requests and scheduled work; Gemini supports classification and copy; the platform document describes a separate Nano Banana image-generation pipeline. The modules share account and store context instead of each becoming a standalone tool.
ONE WORKFLOW
From input to outcome
For a Meta campaign, a seller selects product URLs, creative source, targeting, countries, and budget structure in the dashboard. The server action authenticates the user, resolves the effective account ID, checks feature availability for the campaign batch, and sends pending campaign requests to an n8n webhook. A missing webhook URL or non-success response returns an error; only an accepted request increments feature usage and refreshes the campaign list. The platform document describes downstream creative generation and Meta campaign creation, while the checked-in application shows campaign status and Meta-ID fields for review.
ENGINEERING
Decisions visible in the code
- The application checks account limits before handing work to n8n, so a disallowed request never starts external processing.
- Asynchronous campaign requests separate the seller-facing configuration step from slower creative and API work, while status fields let the dashboard show progress and failures.
- Shopify and Stripe webhook handlers verify provider signatures before acting on external events.
BENEFIT & EVIDENCE
What the workflow enables
A seller can start from existing store context and coordinate creative, campaign, support, and reporting work through one dashboard. This is a functional workflow benefit; the source does not establish measured conversion or staffing gains.
The supplied case-study document describes nine modules and reports speed and automation metrics, but no measurement method or live execution evidence was provided for those figures. The portfolio therefore describes implemented and documented workflows without publishing the numbers.
EXPLORE NEXT