Product Design • UX Research • AI/ML • 3D Technology

HoloSnap

An AI-powered web application that transforms 2D images into interactive 3D models, democratizing 3D content creation for designers, marketers, and e-commerce businesses.

Role
Lead Product Designer
Timeline
6 months
Team Size
12 People
Tools
Figma, Miro, Blender, WebGL

Collaboration & Stakeholders

Collaborated cross-functionally with designers, engineers, and product managers to build relationships with stakeholders and deliver a complex AI-powered product.

ML Engineering Team

Partnered with ML engineers to translate complex AI capabilities into intuitive user controls. Conducted weekly sync meetings to align on model limitations and possibilities.

  • • Participated in model training feedback sessions
  • • Designed error handling for edge cases
  • • Created visualization strategies for 3D previews
Product Management

Aligned design decisions with business objectives and user needs. Facilitated design critiques and sprint planning sessions to prioritize features.

  • • Weekly design reviews and sprint planning
  • • Feature prioritization workshops
  • • User feedback synthesis and roadmap planning
Frontend Engineering

Worked closely with frontend team to ensure pixel-perfect implementation of 3D viewer and real-time preview features using WebGL and Three.js.

  • • Daily standups and pair programming sessions
  • • Design handoff with detailed specifications
  • • Performance optimization feedback loops
QA & User Testing

Collaborated with QA team to develop test cases and coordinate usability testing sessions. Supported user research and usability testing throughout the design process.

  • • Coordinated 5 rounds of usability testing
  • • Bug triage and UX issue prioritization
  • • Cross-browser and device compatibility testing
Cross-Functional Rituals
3x/week
Design Critiques
Daily
Standups & Syncs
Bi-weekly
Handoff Meetings

Research & Empathize

Understanding user needs, pain points, and expectations through comprehensive research and empathy mapping.

Foundational Research

Research Goal

Understand how users create and share 3D models from 2D product photos, and identify friction points across user types.

Context

HoloSnap.ai is an AI-powered 2D → 3D model conversion tool serving:

  • • Freelance 3D artists
  • • Product/furniture studios
  • • E-commerce sellers and agencies

Research Methods

User Interviews

15 in-depth interviews with designers, marketers, and e-commerce professionals

Competitive Analysis

Analyzed 8 existing 2D-to-3D tools and traditional modeling software

Survey Research

186 responses from potential users across various industries

Observational Studies

Shadowed 6 users during their typical workflow for creating 3D content

Secondary Insights

  • Existing tools demand technical expertise or heavy software installs.
  • Main pain points: long render times, inaccurate models, limited free functionality.
  • Competitive gap: few offer short-term free previews or transparent storage periods.

Opportunity

Simplify the 2D → 3D process with instant AI previews and clear upgrade tiers tied to tangible value (editability, storage, support).

Interview Insights

Interview Insights (Synthesized)

Participant TypeKey Insight
Freelance 3D ArtistWants fast model previews without needing CAD tools.
Furniture SellerNeeds multiple uploads + shareable links for website display.
Game DeveloperPrefers editable layers and lighting controls for refinement.
E-commerce MerchantConcerned about file expiry and wants proactive reminders.
Empathy Mapping
User Mindset"I want to…"Feels when…Needs from system
Curious GuestTest HoloSnap without commitmentFile upload failsGentle feedback and retry option
New CreatorVerify output qualityPreview lacks detailClear upgrade call-to-action
Advanced UserManage model edits efficientlyStorage expires suddenlyPredictable renewal reminders
Enterprise LeadEnsure reliability for teamRetouch delayedTransparent SLA + progress visibility

Define

Synthesizing research insights into actionable user personas, journey maps, and clear problem statements.

Affinity Maps

Grouped recurring patterns from interviews and product observations:

Speed & Ease

Quick previews, minimal inputs.

Control & Ownership

Editable models, transparent storage.

Trust & Support

Consistent communication, visible plan benefits.

Scalability

Collaboration tools, enterprise dashboards.

User Personas

Arif

Freelance 3D Artist

Goal

Generate quick 3D previews for client mock-ups.

Needs

  • • Fast conversion
  • • Editable output
  • • Affordable upgrades

Pain

Limited free usage and non-editable exports.

Meera

Furniture Merchant

Goal

Display interactive 3D product renders online.

Needs

  • • Reminders
  • • Easy renewal
  • • Share/embed links

Pain

Unclear storage expiry and renewal process.

Ravi

Enterprise Manager

Goal

Manage bulk uploads with guaranteed turnaround.

Needs

  • • Dedicated support
  • • Integrated monthly reporting

Pain

Manual retouch tracking, SLA confusion.

User Journey Map & Pain Points
StageActionExperiencePain PointOpportunity
AwarenessClicks "Try Free"Curious but uncertainUnsure of qualityAdd live demo preview
ConsiderationUploads imagesGets previewFile-type errors (.heic)Smarter file validation
ConversionAttempts downloadBlocked by signupFrustrationSoft signup with progress save
AdoptionRegisters & uploadsEngagedHits asset limitDynamic upgrade prompts
RetentionUpgrades planContinues workRetouch confusionClear ticket status tracking
RenewalNear expiryNeeds reminderForgets deadlineAuto-renew toggle
Problem Statements

Problem Statement 1

Guests drop off at download due to unclear conversion flow.

Problem Statement 2

Registered users reach hidden limits without understanding benefits of paid tiers.

Problem Statement 3

Paid users face friction during retouch and expiry cycles.

Problem Statement 4

Enterprise clients lack visibility into SLA adherence and monthly analytics.

Goal Statement

Design Goal

Build a friction-less, AI-assisted 3D modeling experience where users can easily upload, preview, and manage 3D assets — progressing from free testing to professional use through transparent, supportive upgrade paths that establish trust and long-term value.

Ideate

Participated in brainstorming and ideation sessions to generate creative solutions for the 2D-to-3D conversion experience.

3.4 Solution Prioritization Matrix (Impact vs Effort)

Evaluated feature ideas based on user impact and implementation effort to guide MVP development and roadmap planning.

Quick Wins

High Impact, Low Effort

  • • Drag-and-drop upload
  • • Real-time progress indicator
  • • Basic 360° viewer
  • • One-click export

Major Projects

High Impact, High Effort

  • • AI model training optimization
  • • Advanced mesh editing tools
  • • AR preview functionality
  • • Batch processing engine

Fill-ins

Low Impact, Low Effort

  • • Custom background colors
  • • Model rotation presets
  • • Social media share buttons
  • • Usage statistics dashboard

Avoid

Low Impact, High Effort

  • • Custom shader editor
  • • Multi-object scene builder
  • • Real-time collaboration
  • • Video-to-3D conversion

MVP Feature Set Decision

Based on the prioritization matrix, we decided to focus on Quick Wins for MVP and phase one Major Project (AI optimization) over 3 months:

8
Quick Win Features
2
Major Projects
4
Fill-in Features

Prototype

Created wireframes, journey maps, and prototypes to communicate design concepts and user flows to stakeholders.

4.1 Wireframes & Mockups

Developed low-fidelity wireframes for key screens to establish information hierarchy and user flow before investing in high-fidelity designs.

Upload Screen Wireframe

Upload & Setup Flow

Focused on making image upload prominent and intuitive with clear CTAs and format requirements.

Processing Screen Wireframe

Processing & Preview

Designed progress indicators and preview states to manage user expectations during AI processing.

3D Viewer Wireframe

3D Viewer Interface

Structured controls around the viewer for rotation, lighting, and zoom without cluttering the viewport.

Export Options Wireframe

Export & Download

Simplified format selection with clear descriptions of each export type and use case.

4.2 Interaction Flows & User Journeys

Mapped detailed interaction flows to ensure smooth transitions and logical progression through the app.

Primary Flow: First-Time User

Landing
Onboarding
Upload
Process
View 3D
Export

Secondary Flow: Returning User

Dashboard
Quick Upload
Batch Process
Review All
Bulk Export

Error Recovery Flow

Error Detected
Show Reason
Suggest Fix
Retry/Replace
4.3 & 4.4 Paper Prototypes & Low-Fidelity Clickable Models

Created both paper prototypes for rapid team feedback and Figma clickable prototypes for early usability testing.

Paper Prototyping Session

  • 3 design sprint sessions with 8-10 team members
  • Tested 5 different navigation patterns
  • Identified optimal button placement through user simulation
  • Cost-effective way to explore alternative layouts

Figma Clickable Prototype

  • 15 interconnected screens simulating full flow
  • Shared with 20+ stakeholders for early feedback
  • Conducted 6 remote usability tests
  • Iterated based on confusion points and drop-offs

Key Learnings from Prototyping

  • Users expected instant preview, not a loading screen—led to instant low-res preview feature
  • Upload button was initially too small—increased size by 40% based on testing
  • 3D viewer controls needed labels, not just icons—improved accessibility and learnability
4.5 Prototype Test Checklist

Established a comprehensive testing checklist to ensure consistent evaluation across all prototype iterations.

Usability Criteria

Can users upload an image without instruction?
Do users understand the processing status?
Can users rotate the 3D model intuitively?
Is the export process clear and straightforward?
Do error messages guide users to resolution?
Are all interactive elements discoverable?

Performance Criteria

Time to complete first conversion: Target < 3 minutes
Number of clicks from upload to export: Target ≤ 5
Error rate during critical tasks: Target < 10%
User satisfaction score (SUS): Target > 75

Design

Documented design decisions and maintained specifications for high-fidelity designs, design system components, and developer handoff.

5.1 High-Fidelity UI Designs

Elevated low-fidelity wireframes to polished, production-ready designs with attention to visual hierarchy, color, typography, and micro-interactions.

Key Design Decisions

Dark Mode First

Designed with dark theme as primary to reduce eye strain during extended 3D viewing sessions and make models stand out with better contrast.

Gradient Accents

Used subtle blue-to-purple gradients for CTAs and progress bars to convey AI-powered innovation while maintaining professional appearance.

Generous Whitespace

Prioritized breathing room around 3D viewer to minimize visual clutter and keep focus on the model. 80/20 split: 80% viewer, 20% controls.

Depth & Elevation

Employed subtle shadows and border treatments to create visual hierarchy without heavy skeuomorphism. 3 elevation levels: background, surface, overlay.

5.2 Design System Components

Contributed to design system and accessibility best practices by creating reusable components with comprehensive documentation.

Core Components

  • Upload Zone

    Drag-drop component with file validation

  • Progress Bar

    Multi-stage indicator with time estimates

  • 3D Viewer Canvas

    WebGL wrapper with consistent controls

  • Export Modal

    Format selector with preview options

Utility Components

  • Toast Notifications

    Success, error, and info variants

  • Empty States

    Contextual illustrations and CTAs

  • Loading Skeletons

    Animated placeholders for async content

  • Tooltips

    Contextual help with smart positioning

5.3 Typography & Color Tokens

Typography System

Inter

Primary font family for UI elements and body text

Display48px / Bold
Heading 132px / Semibold
Heading 224px / Semibold
Body16px / Regular
Caption14px / Medium

Color Palette

Primary Blue
#3B82F6
Accent Purple
#A855F7
Success Green
#22C55E
Error Red
#EF4444
Surface Dark
#1F2937

Accessibility Considerations

  • All color combinations meet WCAG 2.1 AA standards (minimum 4.5:1 contrast ratio)
  • Interactive elements have focus indicators with 3px outline for keyboard navigation
  • Text scales responsively with rem units to respect user font size preferences
5.4 Motion & Micro-Interaction Guidelines

Defined motion principles to create fluid, purposeful animations that enhance usability without distraction.

Easing Functions

Standardease-out
Entercubic-bezier(0, 0, 0.2, 1)
Exitcubic-bezier(0.4, 0, 1, 1)
Bouncecubic-bezier(0.68, -0.55, 0.27, 1.55)

Duration Standards

Micro100-150ms
Short200-300ms
Medium400-500ms
Long600-800ms

Key Micro-Interactions

Button Hover

Scale 1.0 → 1.05 with subtle shadow increase

150ms ease-out
Upload Zone Active

Border pulse animation when file dragged over

300ms cubic-bezier
3D Model Load

Fade in + gentle rotation on first appearance

600ms ease-out
Success Celebration

Confetti burst + scale animation on export

800ms bounce
5.5 Documentation for Developers

Documented design decisions and maintained specifications to ensure pixel-perfect handoff and implementation.

Handoff Package Includes

Figma Dev Mode

Inspect-ready files with auto-generated code snippets

Component Specs

Detailed spacing, sizing, and state documentation

Interaction Notes

Annotated flows with animation timing and triggers

Asset Exports

Optimized SVGs, icons at multiple resolutions

Responsive Breakpoints

Mobile (320px), Tablet (768px), Desktop (1440px)

Accessibility Checklist

ARIA labels, keyboard nav, screen reader guidance

Collaboration Process

  • • Weekly design-dev sync meetings to review implementation progress and address technical constraints
  • • Slack channel for quick clarifications and design QA feedback
  • • Storybook integration for component library documentation and testing
  • • Final design QA pass before each release to ensure fidelity to specifications

Test

Supported user research and usability testing through multiple rounds of testing, measuring and analyzing feature performance.

6.1 Usability Testing Reports

Conducted 5 rounds of moderated usability testing with 30+ participants across target user segments.

Testing Methodology

5
Testing Rounds
32
Total Participants
45min
Session Duration
  • • Think-aloud protocol to capture real-time thoughts and frustrations
  • • Task-based scenarios: upload image, adjust settings, export model
  • • Post-test surveys including System Usability Scale (SUS) questionnaire
  • • Remote testing via Zoom with screen recording for analysis

Key Findings & Iterations

Round 1: Initial Prototype
  • Issue: 60% of users couldn't find export button
  • Fix: Made export button persistent and more prominent
Round 3: Beta Release
  • Issue: Users unsure if upload was successful
  • Fix: Added confirmation toast with upload preview thumbnail
Round 5: Pre-Launch
  • Success: 92% task completion rate achieved
  • Success: SUS score improved from 68 to 82
6.2 A/B Test Results

Measured and analyzed feature performance through controlled A/B experiments with real users post-launch.

Test 1: Upload Button CTA Text

Variant A
Upload Image
34%
Conversion Rate
Variant B (Winner)
Create 3D Model
47%
Conversion Rate

Result: Outcome-focused CTA increased conversion by 38%. Users responded better to value proposition than action description.

Test 2: Progress Indicator Style

Variant A: Spinner
48%
Perceived Speed
Variant B: Progress Bar (Winner)
67% complete
73%
Perceived Speed

Result: Progress bar with percentage made AI processing feel 52% faster. Determinate indicators reduced anxiety during wait time.

6.3-6.6 Post-Launch Analytics & User Feedback

Measured and analyzed feature performance through analytics, error logs, user surveys, and continuous improvement.

Analytics Data (First 3 Months)

Models Created12,437
Avg. Time to First Model4m 12s
Completion Rate89%
Export Success Rate94%

User Satisfaction Survey Results

SUS Score82/100
Would Recommend87%
Ease of Use Rating4.6/5.0
Quality Satisfaction4.2/5.0

Error Logs & Issues Identified

Image upload fails for files > 10MB6% of attempts

Resolution: Implemented client-side compression before upload

3D viewer crashes on mobile Safari3% of mobile users

Resolution: Added WebGL fallback renderer for unsupported browsers

Export timeout for complex models2% of exports

Resolution: Increased timeout threshold and added progress feedback

Improvement Recommendations

  • Add Batch Processing: 42% of users requested ability to process multiple images simultaneously
  • Improve Model Quality Controls: Users want more granular control over mesh density and texture quality
  • Add Social Sharing: 28% of users attempted to share their 3D models on social media
  • Expand Export Formats: Users requested support for additional formats like USD and STL for 3D printing

Outcome & Reflection

Project outcomes, key learnings, metrics, and design impact on users and business objectives.

7.1 & 7.2 Key Outcomes, Metrics & Results
78%
First Model Success Rate

Users created their first 3D model within 5 minutes

92%
Task Completion Rate

For the core 2D-to-3D conversion flow

40%
Retention Increase

After implementing progressive onboarding

Business Impact

  • 12,437 3D models created in first 3 months post-launch
  • 2,800+ active users with 68% monthly retention rate
  • 4.6/5.0 average user rating on product review platforms
  • 30% conversion to paid tier exceeding initial projections by 15%

User Experience Impact

  • 90% reduction in time to create 3D content compared to traditional methods
  • Zero learning curve for 87% of users—no tutorial needed
  • SUS score of 82 (Grade A) indicating excellent usability
  • 94% would recommend to colleagues and friends
7.3 Design Impact Summary & Key Learnings

What Worked Well

  • Progressive Disclosure: Revealing advanced features gradually prevented overwhelm and maintained high task completion rates among novice users.
  • Instant Preview: The low-res preview feature reduced perceived wait time by 60% and gave users confidence to proceed before committing credits.
  • Collaborative Design Process: Daily standups with engineering enabled quick resolution of technical constraints and fostered mutual understanding.
  • Early & Frequent Testing: Five rounds of usability testing caught major issues before launch and validated design decisions with real user data.

Challenges & How We Overcame Them

  • Challenge: AI output quality varied significantly based on input image.
    Solution: Added pre-upload quality check with suggestions for improving source images (lighting, angle, clarity).
  • Challenge: 3D viewer performance on older devices caused crashes.
    Solution: Implemented adaptive quality settings that automatically adjust polygon count based on device capabilities.
  • Challenge: Users confused by file format options during export.
    Solution: Created use-case-based export wizard ("For web embedding" vs "For editing in Blender") instead of technical format names.

Personal Learnings & Growth

  • Deepened expertise in designing for AI/ML products—learned to balance user control with algorithm confidence, manage uncertainty, and set appropriate expectations.
  • Improved skills in 3D interaction design and WebGL performance optimization, expanding beyond traditional 2D interface design.
  • Strengthened ability to advocate for user needs when balancing technical constraints, business requirements, and user experience—successfully negotiated for instant preview feature despite engineering concerns.
  • Developed data-driven decision-making mindset through A/B testing and analytics, moving beyond assumptions to validate design choices with quantitative evidence.

Future Opportunities

Based on user feedback and market trends, we identified several opportunities for future iterations:

  • Batch processing for teams and agencies handling multiple products
  • Advanced editing tools for fine-tuning AI-generated models
  • Collaborative features for team workflows and feedback loops
  • Mobile app for on-the-go 3D creation and AR preview
  • Integration with design tools like Figma, Adobe XD, and Blender

Key Professional Contributions

Skills and activities demonstrated throughout this project

Brainstorming & Ideation

Participated in brainstorming and ideation sessions with cross-functional teams to generate innovative solutions for AI-powered 3D generation.

Design Artifacts

Created wireframes, journey maps, and prototypes to communicate design concepts and user flows to stakeholders and development teams.

Documentation

Documented design decisions and maintained specifications throughout the project lifecycle, ensuring clear handoff to engineering teams.

User Research

Supported user research and usability testing through 15+ interviews, surveys, and iterative testing sessions with target users.

Cross-Functional Collaboration

Collaborated cross-functionally with designers, engineers, and product managers through daily standups, design critiques, and sprint planning.

Design System & Accessibility

Contributed to design system and accessibility best practices, ensuring WCAG 2.1 AA compliance and consistent component usage.