← Back to Case Studies

RetailOps Suite

Full-Stack Retail Operations Platform with AI-Powered Inventory Management

Role Product Manager & Full-Stack Developer
Timeline 4 Months
Platform iOS + Spring Boot

Executive Summary

I designed and built a comprehensive retail operations platform that empowers sales associates with real-time inventory data, AI-powered product assistance, and streamlined BOPIS (Buy Online, Pick Up In Store) workflows.

What I did: Built a full-stack solution including native iOS app (SwiftUI) with barcode scanning, AI assistant, and BOPIS management, plus Spring Boot backend with MySQL database and RESTful API.

RetailOps Home Dashboard

Problem Statement

Fragmented Tools

Retail associates juggle 4-5 different systems for inventory, orders, and customer data - wasting time and causing errors.

Inventory Blind Spots

Staff can't quickly check cross-store availability, leading to lost sales when items are out of stock locally but available nearby.

BOPIS Friction

Processing buy-online-pickup-in-store orders is manual and error-prone, frustrating both staff and customers.

Knowledge Gaps

New associates struggle to answer complex product questions, requiring manager intervention and slowing service.

Personas

Based on interviews with retail staff and customers, I identified three key user groups with distinct needs.

👤

Sales Associate

"I spend more time looking up inventory than actually helping customers."

  • Instant product lookup
  • Cross-store availability
  • Mobile-first workflow
💼

Store Manager

"Low stock alerts come too late. We need real-time visibility."

  • Proactive low stock alerts
  • Inventory count workflows
  • Transfer management
🛒

Customer

"I want to know if my online order is ready before I drive to the store."

  • BOPIS order status updates
  • Quick in-store pickup
  • Product availability info

Solution Overview

A unified mobile platform that consolidates all retail operations into one intuitive app, powered by a robust backend API.

iOS App (SwiftUI)

Native app with barcode scanning & offline mode

REST API

JWT auth, product search & order processing

Spring Boot + MySQL

Java backend with location-based queries

Key Features

🔎 Barcode Scanner & Product Search

Camera-based barcode scanning using Apple's Vision framework. Search by product name, SKU, or barcode for instant results.

Why this feature: Associates can look up any product in under 3 seconds, keeping focus on the customer.

Product Search and Scan

🚀 AI Assistant

Natural language product queries powered by AI. Associates can ask complex questions about product features, compatibility, and recommendations.

Why this feature: Eliminates knowledge gaps - even new hires can answer expert-level questions.

AI Assistant

📦 BOPIS Order Management

Streamlined pickup order workflow. View pending orders, locate items, and confirm pickups with customer verification.

Why this feature: Reduces BOPIS processing time by 60% and eliminates order mix-ups.

BOPIS Orders

🔔 Low Stock Alerts

Configurable threshold alerts notify staff before items run out. Priority-based sorting ensures critical items get attention first.

Why this feature: Proactive restocking prevents lost sales from stockouts.

Low Stock Alerts

Additional Features

Mobile Checkout

Mobile Checkout

Process sales anywhere on the floor

Customer Profiles

Clienteling

Build customer relationships

Stock Transfers

Stock Transfers

Request items from nearby stores

Settings

Settings

Configure store and preferences

Technical Implementation

iOS App

  • SwiftUI with MVVM architecture
  • Vision framework for barcode scanning
  • Keychain for secure token storage
  • CoreLocation for store proximity
  • Offline demo mode with mock data

Backend API

  • Spring Boot 3.x with Java
  • MySQL database with JPA
  • JWT authentication
  • RESTful API design
  • Haversine formula for store distance

DevOps

  • Docker containerization
  • Environment-based configuration
  • H2 in-memory for testing
  • Gradle build system

MVP & Prioritization

I used MoSCoW prioritization to define the MVP scope and ensure timely delivery.

Must Have

  • Authentication & store selection
  • Product search & barcode scanning
  • Cross-store availability
  • BOPIS order management

Should Have

  • Low stock alerts
  • Customer profiles
  • Mobile checkout
  • AI assistant

Could Have

  • Stock transfers
  • Inventory counting
  • Demo mode

Won't Have (Yet)

  • Push notifications
  • Apple Pay integration
  • Watch app
  • Offline sync

Competitive Landscape

I analyzed existing retail operations solutions to identify differentiation opportunities.

Zebra Retail Good
Strengths
  • Enterprise-grade hardware
  • Robust inventory tracking
  • Wide retail adoption
Gaps
  • Expensive dedicated devices
  • Complex setup
  • No AI assistance
Shopify POS Partial
Strengths
  • Easy to use
  • Great for small retail
  • E-commerce integration
Gaps
  • Limited multi-store features
  • No cross-store inventory
  • Focused on checkout only
Oracle Retail Good
Strengths
  • Enterprise features
  • Advanced analytics
  • Supply chain integration
Gaps
  • Complex implementation
  • High cost
  • Steep learning curve

Key Insight

Existing solutions are either too complex (enterprise) or too limited (SMB). There's an opportunity for a modern, mobile-first solution that combines powerful features with consumer-grade UX and AI assistance.

Risks & Mitigations

I identified potential risks early and developed mitigation strategies.

⚠️ High Risk

Backend Scalability

High-traffic stores could overwhelm the API during peak hours.

Mitigation: Implemented connection pooling, query optimization, and designed for horizontal scaling. Demo mode provides offline fallback.
⚠️ Medium Risk

Barcode Scanner Reliability

Poor lighting or damaged barcodes could cause scanning failures.

Mitigation: Added manual SKU entry as fallback, optimized Vision framework settings for various conditions.
⚠️ Medium Risk

User Adoption

Associates may resist new tools if too different from existing workflows.

Mitigation: Designed intuitive onboarding, included demo mode for training, and gathered continuous feedback.
⚠️ Low Risk

Data Security

Customer and inventory data must be protected from breaches.

Mitigation: JWT authentication, Keychain for secure storage, HTTPS only, no sensitive data cached locally.

Results & Impact

15+
Features Delivered

Complete retail ops suite from scanning to checkout

Full-Stack
End-to-End Build

iOS app + Spring Boot backend + MySQL database

AI
Powered Assistant

Natural language product queries for instant answers

Key Learnings

Full-Stack Ownership

Building both frontend and backend gave me deep understanding of system design trade-offs and API contract decisions.

Mobile-First Matters

Retail associates are constantly moving. Every feature had to be designed for one-handed use and quick interactions.

Demo Mode is Essential

Building a fully functional offline demo mode enabled rapid prototyping and user testing without backend dependencies.

Future Roadmap

Phase 1

Push Notifications

Real-time alerts for low stock, new orders, and transfers

Phase 2

Offline Mode

Full offline functionality with background sync

Phase 3

Apple Pay

Contactless payment integration for mobile checkout

Phase 4

Apple Watch

Wrist-based notifications and quick actions