A clearer online buying experience
iScylla Shop
A digital-product shop that brings browsing, payment, order updates, and after-sales support into one place.
Outcome
Customers can see prices and stock, pay by QRIS, and return to their account for receipts, delivery information, and order-related help.
My role
Founder / Full-Stack Engineer
How I helped
- Built the bilingual storefront and customer accounts.
- Connected payment to order tracking.
- Brought product availability, receipts, and support into the buying journey.
Product
The workflow in context
Captured from iscylla.web.id on September 20, 2026.

The solution
Make each step clear before asking customers to pay: what they are buying, how it is delivered, what the warranty covers, and where to get help.
What changed
The shop is live with a populated catalog and customer purchase journey. Sales and conversion measurements are not published.
Lessons learned
A clear buying process and accessible order records help customers understand what happens next.
Behind the solution
Digital-product buyers need clear product terms, delivery methods, warranty information, payment status, and support after checkout instead of an informal chat-only purchase flow.
A Ruby on Rails storefront serves localized catalog and account flows. Product media is delivered through Cloudflare R2, while checkout connects QRIS confirmation to persisted order, receipt, delivery, and support records.
Constraints
- Keep checkout familiar for Indonesian buyers through QRIS and Rupiah pricing.
- Make stock, delivery method, duration, and warranty visible before purchase.
- Keep receipts, delivered access, and support context available in one customer account.
Technical decisions
- Required customer accounts so orders, receipts, and delivered access remain recoverable.
- Connected payment confirmation to explicit order and fulfillment states.
- Kept product media separate from application storage with Cloudflare R2.
Key features
- Bilingual catalog with transparent prices and stock.
- QRIS checkout with order tracking and receipts.
- Delivery information and support tied to each order.
AI system implementation
No AI runtime is documented for this storefront. Catalog, QRIS confirmation, order states, receipts, and fulfillment use application logic and persisted records. Payment and delivery correctness do not depend on model responses.
Could a similar approach help your business?
Tell me where work gets stuck. We can identify a practical first improvement and what success should look like.
Discuss your challenge