Arxec Systems
Back to Archive
Case Study · 2024INTERNAL

Southside Flow

Southside Flow

Restaurant POS, Kanban kitchen display system (KDS), and inventory ledger built with Next.js and Supabase Realtime.

Southside Flow

Restaurant operations suite built with Next.js, Supabase Realtime, and Chart.js. Combines a touch-friendly staff POS (/pos) with add-on modifiers, a 3-column Kanban Kitchen Display System (/kds: New, Preparing, Ready), a Zomato order bridge, start-of-day/end-of-day inventory reconciliation, and streaming CSV analytics exports.

Problem Visual
The Problem

Paper kitchen tickets & disconnected stock counts

Front-of-house staff and kitchen line cooks relied on paper tickets and verbal callouts during peak hours, while daily ingredient prep batches and end-of-day stock counts weren't tied to actual POS order deductions.

Solution Visual
The Solution

Supabase Realtime POS-to-KDS & batch inventory

Engineered a mobile-first POS and Kanban KDS connected via Supabase Realtime subscriptions so kitchen screens update the instant an order is fired. Added a batch inventory ledger (start-day, prep-batch, receive, adjust, EOD) and Chart.js analytics for menu attach rates and peak-hour throughput.

Feature Visual
Implementation

Live Kanban KDS & Inventory Reconciliation

Supabase
Next.js API Routes
WebSockets
Role-Based Access

System Core · Efficient Large Data Export with Generator

async function* getOrdersGenerator(supabase) {
  let rangeStart = 0;
  const BATCH_SIZE = 100;

  while (true) {
    const { data: orders, error } = await supabase
      .from('orders')
      .select(`
        id, order_number, status, total_price, created_at, order_type,
        orderitems (id, quantity, price_at_order, variant_name,
          menuitems (name, category),
          orderitem_addons (addon_id, price_at_order, addons (name)))
      `)
      .order('created_at', { ascending: false })
      .range(rangeStart, rangeStart + BATCH_SIZE - 1);

    if (error) throw error;
    if (!orders || orders.length === 0) break;

    yield orders;
    
    if (orders.length < BATCH_SIZE) break;
    rangeStart += BATCH_SIZE;
  }
}
javascript
Next Project

Next Case Study

Ashly