Synapse

Synapse

Role

Research

UIUX design

Timeline

Jan-Apr 2026

Team

Solo design

Tools

Lovable

Figma Make

Figma

65% of designers & developers, spend 4-8 hours weekly
All stuck explaining & Interpreting the same thing.

The Problem

The Translation Tax

Teams lose 20–30% of working hours re-explaining the same decisions across tools, roles, and meetings. Context exists, it's just trapped in the wrong format, in the wrong place.

The Solution

A coordination layer that integrated across the cross-functional ecosystem—from design canvases and communication hubs to the developer stack. By transforming ambient team intent into technical action, Synapse eliminates the manual documentation tax to ensure design vision and production never drift apart, improving efficiency.

Handoff flow: Designer View

Smart handoff in one click

Handoff flow: Engineer View

Implement with full context

Conflict flow: Product Manager & Designer's View

Autonomous detection & real time communication

The Process

It Started From My Internship Experience
01
🔗
Designer
Competitive Handoff
Relied purely on manual handoff, endless communication is a burden.
Design building
02
😳
Engineer
Built the wrong thing.
Unable to understand handoff note all the time, or why it's designed that way.
Sprint Review
03
💬
PM
One message. Never synced.
Hard to keep track, Figma might not hear the message in slack in real time.
3 days earlier · Slack
04
🔁
All roles
Every sprint. Same story.
Three tools. Zero shared memory.
Pattern
This is not a single case, but an ongoing common pain

Primary Research of the collaboration problems that each role face

Major Painpoints
01

Tools don't speak to each other, the same context gets re-explained across platforms, which makes coordination become manual, repetitive, and lossy.

02

Design and code drift silently. By the time the conflict surfaces, it's already mid-sprint and expensive to fix.

03

Every role is working from a different version of reality, no shared source of truth, no audit trail, no closure.

01

Tools don't speak to each other, the same context gets re-explained across platforms, which makes coordination become manual, repetitive, and lossy.

02

Design and code drift silently. By the time the conflict surfaces, it's already mid-sprint and expensive to fix.

03

Every role is working from a different version of reality, no shared source of truth, no audit trail, no closure.

01

Tools don't speak to each other, the same context gets re-explained across platforms, which makes coordination become manual, repetitive, and lossy.

02

Design and code drift silently. By the time the conflict surfaces, it's already mid-sprint and expensive to fix.

03

Every role is working from a different version of reality, no shared source of truth, no audit trail, no closure.

User Types

UX

Designers

Primary

PAIN POINT

Spends hours re-explaining design decisions to engineers and PMs.

manual burden of writing redlines and technical specs for every screen state.

Lacks context on system constraints.

Conversation of one design task happen everywhere

WHAT THEY NEED

Effortless Documentation

On-demand rationale generator

Dialogue to design guide

Constraint reminders

SE

Software Engineer

Primary

PAIN POINT

Struggle to understand the logic for interactions

Design sometime uses non-standard tokens that aren't in the library

WHAT THEY NEED

Visual diff since last implementation

Conflict detail cards

Version constraint flags

Linked Figma specs in GitHub

PM

Product Manager

Secondary

PAIN POINT

Loses track of design decisions

Manually writes tickets from meetings

Decision Loss

WHAT THEY NEED

Status Assembly

Action Item Tracking

Version-compatibility scan

Competitive Analysis

Not a handoff tools, but a decision & Collaboration lifecycle

The design-to-development handoff market answers "what changed," and require human initiated action.
Synapse is the only tool spontaneously answering "why it changed, who decided, and how it reaches everyone."

Signal Capture Model

Monitors artifacts, not people

Tier 1

Passive trace

Always on — silent

Records artifact events with no UI. Figma, Github saves, Slack messages in opted-in channels, meeting uploads. User is unaware it is running.

Automatic

Tier 2

Proactive badge

Non-interruptive

When an event is detected on the artifact the user is currently viewing, a small dot appears on the Copilot icon. No sound. No popup. User can ignore it.

Opted-in channels only

Tier 3

Summoned copilot

On demand

User opens the Copilot Panel via shortcut or icon click. Reviews suggestions, conflicts, drafts. Every output requires explicit approval before any action is taken.

User-controlled

Design Goals
01

Respect where people already work, reduce coordination friction to improve efficiency without adding a new tool

02

Surface the right information at the right stage

03

Make collaboration visible without making it loud, close the loop across all roles

Design goals

Embed cancer education into people’s daily environments to normalize screening and reduce fear.

Enable individuals to understand their own risk and take first-step actions independently.

Redistribute responsibility from physicians alone to the broader service ecosystem supports users across the full journey in multiple-touch point.

Data Model

How synapse work

Key Design Decision 1

Passive

Proactive

  • Appear only when it has something relevant to say, then recedes

  • panel's content, actions, and urgency shift based on detected workflow state

Key Design Decision 2

Why user autonomy?

  • When the system decides automatically, the user can't predict what will change, can't review it before it happens,

  • Human are professionally accountable for what gets shipped. Auto-sync removes that accountability boundary.

Design System
Reflection
What I learned

55.1% of adults in Los Angeles speak languages other than English at home.

  • Giving AI more capability isn't the hard part. Knowing when to surface it, and when to stay out of the way, is the actual design problem

55.1% of adults in Los Angeles speak languages other than English at home.

  • The user and the buyer are different people. The ROI case to the buyer is time lost to re-explanation and misaligned handoffs, not the panel UX itself.

55.1% of adults in Los Angeles speak languages other than English at home.

What I'd push in the future

55.1% of adults in Los Angeles speak languages other than English at home.

  • Everything I designed still mostly centers the designer's perspective. Push more on PM and engineer side after more testings.

55.1% of adults in Los Angeles speak languages other than English at home.

  • Integrate team-level view. A manager can't currently see where handoffs are stalling across multiple projects.

55.1% of adults in Los Angeles speak languages other than English at home.