HARMONY HACKATHON 2026
From Noise to
Nervous System
How two squads built a working feedback loop for 30,000 store employees — in two days.
FEBRUARY 2026 · AMSTERDAM
H
AGENDA
Today's Journey
01
INTRODUCTION

The Mission & Build Model

Our goal, the 2-squad structure, and the AI-assisted Vibe Coding approach.

02
RESEARCH

Wall of Signal

6 markets interviewed. The diagnosis: friction, filtering, and desire for agency.

03
DIAGNOSIS

Pain Points & Shadow Channels

Where feedback dies and the unofficial paths users take.

04
PITCH 1 — TEAM 1

Killing "High Friction"

The Direct Line: in-app feedback with multimedia, voice notes, and offline support.

05
PITCH 2 — TEAM 2

Enabling "Agency"

The Voice: democratized voting, localization, and duplicate detection.

06
CLOSING

The Ask

Summary, side-by-side comparison, and our ask for Q2 prioritization.

INTRODUCTION
The Mission

We set out to design a "nervous system" for Harmony — to move from noise to actionable signal for our 30,000 users across all markets Harmony operates.


Every feature you're about to see traces back directly to a real user signal from our Global Operations market interviews and the 15 European Digital champions.

30K

Global Users

6

Markets Interviewed

2

Days to Build

INTRODUCTION
The Build Model

12–14 participants split into two cross-functional squads — each with a Designer, Front-end, Back-end, QA, and Product member.


We used an AI-assisted "Vibe Coding" model to build working SwiftUI prototypes in two days. No Figma sketches — real, running code in Xcode.

COPILOT
CLAUDE
VS CODE
XCODE
SWIFTUI
RESEARCH
The Wall of Signal
Before the hackathon, we interviewed 6 Global Operations markets to build a "Wall of Signal" — documenting where feedback dies, shadow channels exist, and what users actually want.
WHAT WE HEARD
🇺🇸
North America
Context loss, device restrictions
🇧🇷
Latin America
WhatsApp as primary feedback tool
🇪🇺
Europe
Broken support, trust destroyed
🇰🇷
Korea
Cultural silence, language barriers
🇯🇵
Japan
Shared Excel workarounds
🌍
Emerging Markets
Gatekeeper filtering
3 clear patterns emerged
Let's look at the diagnosis…
RESEARCH
The Diagnosis
After interviewing 6 markets across 4 continents, three clear patterns emerged — these are the findings that shaped what we built.

🔴 High Friction

Users are forced to leave the app to report issues. Store devices block screen recording. Feedback arrives as vague emotion — not actionable data.

"They describe the problem but we can't see what they're seeing."

🔴 High Filtering

District Managers, CX, and Global Ops manually block signals before they reach product. Cultural silence in Korea. Broken 123-support in Europe destroyed trust.

"Why report it? Nothing happens anyway."

🟢 Desire for Agency

Users want to vote on features, solve their own problems through FAQs, communicate in their own language, and track feedback status.

"The real feedback comes through WhatsApp, not the official tool."
HACKATHON
The Teams
We originally scoped 3 challenge topics. Due to fewer participants than expected, teams voted and picked their challenge — leaving one on the table.
1
TEAM 1 — PICKED ✓

How do we enable a store employee to report a bug or share an idea without ever leaving Harmony — including visual evidence — while physically working?

2
TEAM 2 — PICKED ✓

How do we enable store employees to shape Harmony directly — vote on what matters, see what others need, track what happens — in their own language, without a gatekeeper?

3
TEAM 3 — NOT STAFFED

How do we enable store employees to solve their own Harmony problems instantly — without creating a ticket, without waiting for support — in their local language?

PITCH 1 — TEAM 1
Killing "High Friction"
The Direct Line — Enable a store employee to report a bug or share an idea without ever leaving Harmony.

THE SIGNAL

  • 📱
    Users are forced to leave Harmony to report issues — no native feedback path exists
  • 🔒
    SAM devices block screen recording — staff use personal phones as workaround
  • 💬
    Shadow channels: WhatsApp (LATAM), KakaoTalk (Korea), personal phones (Japan)
  • 😶
    Feedback arrives as vague emotion instead of actionable technical data
1
TEAM ONE
1
PITCH 1 — LIVE DEMO
The Proposal
MULTIMEDIA

Screenshot & Annotation

Auto-capture the current screen, then annotate with PencilKit — draw, highlight, undo/redo. Attached as visual evidence.

VOICE

Voice-to-Text Notes

Real-time speech recognition via SFSpeechRecognizer. Speak your feedback while holding a scanner or box.

CONTEXT

Auto-Captured Metadata

Screen name, device model, app version, store ID, OS version, timestamp — all captured automatically. Zero effort.

TRACKING

Ticket Lifecycle

Human-readable ID (HF-00042), status timeline, and a "Nudge" button for stalled tickets (>3 days unchanged).

SELF-SERVICE

Knowledge Base

Searchable, categorized articles with pinned content and markdown rendering. 7 seed articles ship offline.

ACCESS

Quick Actions Panel

Edge-swipe gesture or 5-tap header opens an instant drawer — Report Bug, General Feedback, Suggest Enhancement, or browse KB.

Team 1 Demo
1
PITCH 1 — CONSTRAINTS
Survives the Store Floor
Built for real-world retail conditions — not ideal lab environments.
📡

Offline-First

All data persisted locally first. Network monitor auto-syncs when connectivity resumes. Unsynced items show a clear indicator. Graceful degradation — never blocks the user.

👍

One-Thumb Operation

Designed for use while holding a scanner or box. Edge-swipe gesture opens the Quick Actions panel. Voice notes eliminate typing. Large tap targets throughout.

📸

No Screen Recording Needed

Auto-captures the current screen as a screenshot programmatically — bypasses SAM device recording restrictions entirely. Annotate with PencilKit overlay.

0

Apps to leave

< 30s

Time to report

100%

Context captured

2
PITCH 2 — TEAM 2
Enabling "Agency" &
Killing "High Filtering"
The Voice — Empower employees to shape Harmony directly, without a gatekeeper, in their own language.

THE SIGNAL

  • 🚫
    A human filter layer (DMs / CX / Global Ops) manually blocks signals before product sees them
  • 🤐
    "Cultural Silence" in Korea & Japan — staff uncomfortable complaining in English. Silence ≠ satisfaction
  • 😤
    Broken 123-support in Europe destroyed trust — users stopped reporting entirely
  • 🗳️
    Europe explicitly requested democratized voting — let popular requests rise naturally
2
TEAM TWO
2
PITCH 2 — LIVE DEMO
The Proposal
VOTING

Democratized Upvoting

Toggle-based voting on any feedback item. Vote counts drive the "Top Requests" ranking — popular requests rise naturally without gatekeepers.

LOCALIZATION

Multi-Language Translation

Each item stores original language + translated text. One-tap "Show Original / Show Translation" toggle. Korean & Japanese input supported natively.

SMART

Duplicate Detection

Keyword-overlap algorithm (40% threshold) surfaces similar existing feedback before publishing. Vote on duplicates instead of creating noise.

FILTERING

Process-Based Filters

Filter by Store Operating Process — Replenishment, Salesfloor, Order Fulfilment, Bins, Endless Aisle, Tasks. Color-coded chips with badge counts.

SUBMISSION

Multi-Step Wizard

"New Idea" or "Report an Issue" with title, description, process tag, and up to 10 photo/video attachments. Anonymous option available.

TRUST

Voice of Stores Hub

"Top Requests" + "My Submissions" tabs. See your feedback, track votes, know that someone is listening and acting.

Team 2 Demo
2
PITCH 2 — CONSTRAINTS
Transcends Barriers
Designed to overcome language barriers, cultural hesitation, and gatekeeper filtering.
🌐

Language-Agnostic

Submit in Korean, read in English. The translation toggle ensures every voice is heard regardless of language. No more "Cultural Silence."

🙈

Anonymous by Default

All submissions are anonymous — removing the social pressure that prevents junior staff from speaking up, especially in hierarchical cultures.

🔄

Passive Signal Capture

Even without formal complaints, the voting mechanism captures "passive signal" — a vote is a low-friction agreement that a problem matters.

0

Gatekeepers

Languages supported

1 tap

To vote / translate

SUMMARY
Two Squads, One Nervous System
TEAM 1 — THE DIRECT LINE

Kill High Friction

In-app feedback with screenshot annotation, voice-to-text, auto-captured metadata, ticket tracking with nudge, and a full knowledge base — all offline-first and one-thumb operable.

Signals addressed: No direct channel · Context loss · Shadow channels · Excel workarounds · Device restrictions
TEAM 2 — THE VOICE

Kill Filtering · Enable Agency

Voice of Stores with democratized voting, multi-language translation, duplicate detection, process-based filtering, anonymous submission, and a top requests hub.

Signals addressed: Gatekeeper filtering · Cultural silence · Language barriers · No status visibility · Trust destruction
THE ASK
These Are Not Slide Decks.
These Are Working Prototypes.
Built in SwiftUI. Running in Xcode. Backed by Supabase.
We ask that these two proposals feed directly into the upcoming Q2 prioritization discussions, with the goal of selecting a concept for a Q3 Pilot.

Today

Working prototypes demonstrated to Dharma, Niall, and Vandana

Q2 Prioritization

Feed proposals into discussion. Evaluate build vs. buy. Technical feasibility review.

Q3 Pilot

Select concept. Run pilot in 1–2 markets. Measure signal capture improvement.

Thank You
From noise → to nervous system.
Let's give our 30,000 users a voice.
HARMONY HACKATHON · FEBRUARY 2026 · AMSTERDAM