Inherited 60,000 lines of undocumented React Native and a security vulnerability exposing every user's photos. Fixed it in week one, shipped Phase 1 inside the legacy code, then led the full native iOS rebuild — 581 Swift files, 104K LOC, 303 SwiftUI views.
Timeline
Sep 2024 → present · Native rebuild Nov 2025
Role
Lead Engineer & Designer
Technology
Swift, SwiftUI, Firebase, Cloud Functions
Platform
iOS native
5,000+
Production users
104K LOC
Native Swift codebase
15× zoom
Coordinate-locked precision
The Redesign
Before and after.
Same product, two codebases — the legacy React Native build beside the native SwiftUI rewrite.
BeforeLegacy · React Native
AfterNative · SwiftUI
The Product
The Product
HornScore is a hunting app for serious hunters. Users photograph trail-cam footage, field-judge antler measurements using anatomical references, calculate official scores adapted from Boone & Crockett methodology, share scores with a community, and access species-specific guides.
The audience invests in premium equipment and expects the software to match. This is a production app with 5,000+ users.
Screens
Eighty-seven screens, one design system.
Every flow in the production app — scoring, filters, settings, social, guides, messaging, comments, albums, reports — built from a single tokenized system. A representative selection is shown below.
What I Inherited
Week one: audit findings.
Before writing a single feature, I documented what was actually in the repository. Here's the report I wrote to the founders.
audit findings · week 01
- 0150,000–60,000 lines of React Native
- 02Three partially implemented versions of the app, in parallel
- 03No versioning. No tests. No documentation.
- 04Individual files averaged 2,000–5,000 lines each
- 05Firebase storage bucket was publicly readableCRITICAL
→ The security finding
Every user's trail-cam photos, hunting locations, and personal images were accessible to anyone with the URL. I documented the exposure, implemented proper Firestore rules, and verified the fix before touching anything else.
Phase 1
Phase 1 — ship inside what already exists.
Before proposing a rebuild, I shipped the Line Reference Tool inside the legacy React Native codebase. The goal was simple: prove that complex new functionality could be delivered in the existing code, on time, without breaking anything.
It was the work that earned the trust to propose the bigger move.
Line Reference Tool, shipped inside the legacy code.
Phase 2
Phase 2 — the native rebuild.
First commit on the rebuild branch: November 6, 2025. Native Swift, SwiftUI, MVVM + Coordinator architecture, dependency-injection container, repository pattern, design system built to scale. Owned by one person.
581
Swift files
across the repo
104K
Lines of Swift
production code
303
SwiftUI views
156 feature-specific
39
Domain entities
modeled and validated
87
User-facing screens
across 12 feature modules
5
Backend integrations
Firebase, RC, Amplitude, Sentry, FB
Risk assessment from the shareholder progress report: technical risk LOW. The remaining work is feature implementation, not re-architecture.
The Math
The math underneath the measurement tool.
The Line Reference Tool lets users place Bézier curve points on a photo and trace anatomical landmarks at up to 15× zoom — with coordinate-locked precision, even as the user pans, zooms, and re-zooms.
The trick is two bidirectional transforms between canvas space (image pixels) and screen space (the rendered SwiftUI view). Every touch event, every drawn point, every hit-test runs through these:
FieldJudgeScoringViewModel.swift — the coordinate-space math core (2,089 lines).
swift
func screenToCanvas(_ screenPoint: CGPoint, viewSize: CGSize,
totalScale: CGFloat, totalOffset: CGSize) -> CGPoint {
let center = CGPoint(x: viewSize.width / 2, y: viewSize.height / 2)
return CGPoint(
x: (screenPoint.x - center.x - totalOffset.width) / totalScale + center.x,
y: (screenPoint.y - center.y - totalOffset.height) / totalScale + center.y
)
}↳ DetailThe Bézier rendering uses a Catmull-Rom spline with arc-length computed by parametric subdivision (200 intervals per segment), integrated in real time as the user adjusts points. Hit-testing dynamically adjusts a 44pt tap threshold to screen space, and stroke width scales inversely with zoom so lines stay visible without thickening.
The Unified Tool
Two tools became one.
The legacy app had separate straight-line and curved-line measurement modes — two interaction models the user had to keep in their head.
The rebuild collapses both into a single canvas state machine. The first two points calibrate against an anatomical reference; three or more extend into a curved measurement. The transition is automatic; users never see a mode toggle.
The code is just as direct: a seven-state enum, an auto-advancing transition function, and comments throughout marked `// Legacy — should not be entered in unified flow`.
Design System
A 586-line design system, before a single feature shipped.
I built the tokens first. Adaptive light/dark, semantic surfaces, a brand color that adjusts to color scheme, spring curves that match native iOS sheet snap. Every screen in the app draws from this file.
DesignSystem.swift · 586 lines
01 · Color
50+ tokens
RGB(0.745, 0.518, 0.365)
Brand primary
#065767
Brand accent · Light
#1E92A8
Brand accent · Dark
Adaptive
Surface · Dark
02 · Type
20+ sizes
Display 64HornScore
Title 32Field judge accurately.
Body 18Score trophies the right way.
Caption 12ANTLER POINTS · 0.12EM
03 · Spacing
25 tokens · 4pt grid
4pt
8pt
12pt
16pt
20pt
24pt
32pt
48pt
64pt
50+
Color tokens
25
Spacing tokens
20+
Type sizes
40+
Frame sizes
DesignSystem.swift — 586 lines, dark-mode-aware throughout, used across 165 production files.
The Stack
Production infrastructure, wired before the UI.
Auth, analytics, crash reporting, subscriptions — all integrated and verified before a single feature screen was built. Standard practice in mature engineering organizations; rare in solo work.
Firebase
Auth, Firestore, Storage
Cloud Functions
TypeScript · FCM, provisioning
RevenueCat
Subscription billing
A
Amplitude
Product analytics
Sentry
Crash & error reporting
Sign in with Apple
Nonce-based OAuth
Google
OAuth · auto-login
Facebook
OAuth · email verification
Why this matters
Same problem, different vertical.
If your business runs on an iPhone app that captures field data and syncs to a server, this is the same problem shape. The track record I'd bring: audit your codebase for security and architecture risk in week one, ship value inside the existing code while we plan, and only propose a rebuild when the data — not the engineer — calls for it.
william@piehsoft.com →