Navy — AI-Powered Cross-Platform Mobile Application
Problem
A stealth startup needed a production-grade mobile application that could handle complex user interactions, secure payments, and real-time AI-driven features, while maintaining a rapid development cycle across iOS and Android.
Outcome
Delivered a serverless, automatically deployed application with segregated CI/CD pipelines per platform, reducing release friction. The system serves two distinct user types (organisation accounts and personal accounts) with context-aware AI responses, all through highly secure Lambda endpoints. The client is projecting several million GBP in revenue over five years.
Code Sample — Automated CI/CD with Expo EAS
The CI/CD pipeline uses commit-message triggers to selectively build for iOS, Android, or both, and automatically submits to app stores from the main branch:
# .github/workflows/eas-build.yml (excerpt)
jobs:
build_ios:
name: Build iOS
if: ${{ contains(github.commit_message, '[ios]') || contains(github.commit_message, '[native]') }}
environment: ${{ github.ref_name == 'main' && 'production' || 'preview' }}
type: build
params:
platform: ios
profile: ${{ github.ref_name == 'main' && 'production' || 'preview' }}
build_android:
name: Build Android
if: ${{ contains(github.commit_message, '[android]') || contains(github.commit_message, '[native]') }}
environment: ${{ github.ref_name == 'main' && 'production' || 'preview' }}
type: build
params:
platform: android
profile: ${{ github.ref_name == 'main' && 'production' || 'preview' }}
submit_android:
name: Submit Android Build
needs: [build_android]
type: submit
params:
build_id: ${{ needs.build_android.outputs.build_id }}
profile: ${{ github.ref_name == 'main' && 'production' || 'preview' }}
submit_ios:
name: Submit iOS Build
needs: [build_ios]
type: submit
params:
build_id: ${{ needs.build_ios.outputs.build_id }}
profile: ${{ github.ref_name == 'main' && 'production' || 'preview' }}