Case study · Retail & E-commerce · 2025

A regional retailer

Designed and shipped a headless storefront and inventory dashboard, replacing a brittle legacy system and giving the team real-time visibility across branches.

The challenge

A brittle legacy platform coupled the storefront to inventory, so traffic spikes took the whole system down and the team had no real-time view across branches.

What we did

  • Shipped a headless, server-rendered storefront for speed and SEO, decoupled from the inventory backend.
  • Built a real-time inventory dashboard giving staff a single view across every branch.
  • Containerized the stack and set up CI/CD with monitoring, so releases and traffic spikes stopped being scary.
  • Modeled the data for growth — clean APIs the team can build new channels on.

Outcomes

99.9%
uptime
-60%
page load time

Services

Web EngineeringDevOpsBackend Engineering

Stack

TanStack StartNode.jsPostgresDocker
Start a project

Key engineering decisions

The choices that shaped this build — and the reasoning behind them. It’s how we think about a problem, not just what we shipped.

Architecture

TanStack Start for a data-dense operations dashboard

The heart of this build is an app-shell dashboard — dozens of live inventory views, filters, and cross-branch drill-downs. We chose TanStack Start for its first-class routing and fine-grained data fetching + caching: the UI stays instant as staff move between branches, with far less bespoke state plumbing than an SSR-first framework would need for this kind of interactive tool.

-60% page load time
Reliability

Storefront decoupled from inventory

The legacy system took the whole platform down under traffic spikes because the storefront and inventory shared one process. We split them into independent services with their own scaling and failure boundaries, so a surge in shoppers can never take the back-office offline — and vice versa.

99.9% uptime
Performance

Real-time stock without hammering the database

Cross-branch stock updates are pushed and cached at the edge of the data layer rather than re-queried on every view, so staff see near-real-time numbers without the dashboard generating punishing read load as it scales.

Reliability

Containerized delivery with monitoring

The whole stack ships as containers through CI/CD with health checks and monitoring, so releases are routine and traffic spikes are observable instead of frightening. Clean, versioned APIs leave room to add new sales channels without re-platforming.

Let's build

Have something to build?

Tell us what you're working on. We'll get back to you with a clear, honest take on how we can help.