Case study 02
Agents · Browser automation
Dynamic Shopify Storefronts
A Shopify landing-page personalization product that imports Meta ad creative and audience context, captures the existing product page with Playwright, proposes page changes with AI, gives merchants an interactive preview, and publishes approved variants through Shopify app extensions and metafields.
My role
Built the ad-data integration, multimodal analysis, browser capture, agentic planning, interactive preview, safe publishing, dynamic runtime, experimentation, and analytics systems.
Core technologies
System path
A simplified view of the product workflow.
- 01
Meta ad context
- 02
Shopify context
- 03
Browser capture
- 04
Page agent
- 05
Merchant preview
- 06
Publish + attribute
01
Product overview
Ad-specific Shopify landing pages
A shopper may click an ad built around a specific audience, promise, product angle, or creative, then land on the same generic product page as everyone else. The system preserves that context and uses it to produce a more relevant storefront variant.
02
My work
Page capture, AI editing, and merchant review
- 01Ingests campaign, audience, image, video, copy, and performance context from Meta Ads.
- 02Combines Shopify product, page, brand, and design data with merchant instructions.
- 03Captures visible mobile DOM, text, and screenshots through browser automation.
- 04Plans and applies copy, media, layout, and product-page changes.
- 05Presents side-by-side and interactive iframe previews for merchant review.
03
Publishing
Safe publishing and experiment attribution
Selector safety, visual verification, saved states, and health checks reduce the risk of modifying live storefronts. Approved variants publish through Shopify app extensions and metafields rather than brittle theme rewrites.
Visitor assignment stays persistent, while Shopify Web Pixels connect page views, add-to-cart events, checkouts, and purchases back to each experiment variant.