ए

THIS CHEF
WEARS MANY HATS

यह शेफ कई टोपियाँ पहनती है

Hi, I'm Esha - a UX & Product Designer who thinks in systems, feels in aesthetics, and will absolutely judge a product by its empty state screen. Born and raised in India, I bring a maximalist's heart to every minimal interaction. I research like a detective, ideate like a chef improvising with whatever's in the fridge, and design like someone who genuinely cares whether you enjoyed the meal.

👨‍🍳 Researcher 🎨 Visual Designer 🔍 Problem Solver ✨ Ideator 🍽️ UX Designer

DOWNLOAD RESUME LET'S TALK

THE STORY BEHIND THE MENU

WHY THIS.
WHY ME.

🐕

Why is the logo a dog?

A black dog is considered a wayfinder - navigating complexity through intuitive, human-centred design, grounded in a deep commitment to user advocacy. Loyal, sharp, always sniffing out the real problem.

🪔

Why Desi Maximalism?

I'm obsessed with Bollywood posters, spice-tin typography, and the organised chaos of a Mumbai market. My designs are minimal, but I am emphatically not - from the clothes I wear to the art I collect. Desi maximalism felt like the only honest aesthetic for a portfolio that's supposed to be about me.

🍛

Why a Food Menu?

I'm a massive foodie - and I genuinely believe the design process mirrors cooking at a restaurant. You start with a brief (the order), research your ingredients (users + context), test your recipe (prototypes), and plate it beautifully (ship). Both require obsessive attention to the person on the other side of the table.

ESHA MORE

ईशा मोरे

UXD ✦ MFA

New York, NY

काम / WORK

01 · Healthcare · AI/ML · UX Design

ALTASPARQ
AWARE

An AI Tool for Facial Emotion Analysis

Redesigned an emotion-analysis AI platform to help users better understand emotions - transforming complex, data-heavy outputs into clear, scannable, actionable insights.

RoleUX Designer

InstitutionRutgers University

Duration14 Weeks

Team4 Members

30+

Emotional States
Detected in Real-Time

Design Pillars:
Consistency · Navigation · Clarity

3→4/5

Usability Rating
Post Redesign

WCAG AA

Accessibility Target
Met Fully

Problem & Solution / समस्या और समाधान

BEFORE &
AFTER

पहले और बाद में

BEFORE

The Problem

The existing dashboard presented highly technical, AI-driven emotional data that was difficult to interpret quickly - visually cluttered, lacking clear hierarchy, and not actionable for real-time decision-making.

❌ Difficult to interpret quickly
❌ Visually cluttered and overwhelming
❌ Lacking clear hierarchy and prioritisation
❌ Not actionable for real-time decisions
❌ No plain-language summaries
❌ Broken onboarding - FAQ returns 404

AFTER

The Solution

Simplified the dashboard by converting complex analytics into intuitive, scannable UI components - designing clear visual hierarchies and introducing actionable insight summaries instead of raw data.

✅ Intuitive, scannable UI components
✅ Clear visual hierarchies and modular layouts
✅ Actionable insight summaries replace raw data
✅ Plain-language labels replace jargon
✅ Persistent consent indicators built in
✅ WCAG 2.1 AA accessibility compliance

UI Screens / यूज़र इंटरफ़ेस

THE FINISHED
PRODUCT

तैयार उत्पाद

🎬 PROTOTYPE WALKTHROUGH

Dashboard

→

Live Session

→

Analytics

→

Patient Profile

→

Patient Directory

→

Clinical Schedule

Dashboard - Overview

1 / 8

Design System / डिज़ाइन सिस्टम

ALTASPARQ DESIGN SYSTEM

नींव और घटक · Atomic Design Structure

Scroll to explore →→→

TOKENS

Foundation - raw values that everything else references

Colour Tokens

#1B3FD8Primary

#0D2494Deep Blue

#10B981Stable

#EF4444Critical

#F59E0BWarning

#F8FAFCSurface

#1E293BNav Dark

#64748BMuted

Emotion Colour Map

😄
Joy#F2C94C

😢
Sad#2D9CDB

😡
Angry#EB5757

😲
Surprise#F2994A

😨
Fear#9B51E0

🤢
Disgust#27AE60

😐
Neutral#9AA0A6

Spacing Scale

4px · xs

8px · sm

16px · md

24px · lg

32px · xl

48px · 2xl

Border Radius

4px

8px

12px

Full

STYLES

Typography, shadows, grid - visual language rules

Type Scale

Display · 28–30px Bold

Page titles, patient names

Heading · 22–24px Semibold

Panel headings, labels

84%

KPI · 32px Bold

Primary metric values

Body · 14–16px Regular

Notes, descriptions

Caption · 11px Uppercase

Labels, tooltips, tags

Elevation

Level 1 · Cards 0 1px 3px

Level 2 · Dropdowns 0 4px 14px

Level 3 · Modals 0 12px 32px

Grid · 12 Columns

16px gutter · responsive breakpoints

ATOMS

Smallest reusable building blocks

Buttons

Status Badges

● STABLE ● CRITICAL ● FLUCTUATING ● NEUTRAL

Input States

Search patients...

Elena Rodriguez

Invalid patient ID

Toggle / Switch

Live Analysis On

Dark Mode Off

MOLECULES

Atoms combined into functional UI patterns

KPI Metric Card

AVG. ENGAGEMENT

84.2%

↑ +5% vs last month

Emotion Bar Row

😐 Neutral58%

😄 Happy24%

😲 Surprise12%

😡 Angry4%

Biometric Rings

84%

ATTENTION

42%

AROUSAL

68%

VALENCE

ORGANISMS

Full panels and complex layout sections

Navigation Sidebar

ALTASPARQ

📊 Overview

📈 Analytics

🕐 History

📅 Calendar

👥 Patients

⚙️ Settings

Session Table

DATEDUR.EMOTIONSCORE

Oct 2445:12 ● Serene 9.2

Oct 1752:00 ● Distressed 7.8

Oct 1048:30 ● Fluctuating 8.5

Live Insight Card

💡 LIVE INSIGHT

Engagement dropping - try asking an open question to re-engage participants.

UI COMPONENTS

Full feature patterns assembled for production

Toast Notifications

✅

Session Report Generated

Adam Brown · Oct 24 · PDF ready

⚠️

High Distress Detected

Arthur Sterling · Consider check-in

🔔

Session Starting in 5 min

Elena Vasquez · 2:00 PM Today

Scenario Dropdown

Default

Education

Energy & Engagement

Serenity & Connection

✓ Therapy

Consent Banner

🔴 Emotion Analysis Active
Data visible to host only

Confirm Modal

End Session?

This will stop live analysis and save session data. Cannot be undone.

01 TOKENS 02 STYLES 03 ATOMS 04 MOLECULES 05 ORGANISMS 06 UI COMPONENTS

Design Process / प्रक्रिया

HOW THE
DISH WAS COOKED

कैसे बना यह

🔍

DISCOVER

5 in-depth professional interviews
Heuristic eval - Nielsen's 10 heuristics
Competitive analysis: Affectiva, FaceReader
SWOT analysis of platform strengths
Literature review - Affective Computing
Session recording analysis (FullStory)

🗺️

DEFINE

4 core pain points mapped and prioritised
User personas: Educator, Clinician, Researcher
Journey map - Educator's redesigned experience
3 design pillars established
Design evaluation metrics framework
Success targets defined (SUS, NPS, WCAG)

✏️

DESIGN

Low-fi prototype - modular dashboard concepts
A/B tested 2 layout alternatives
High-fi Figma - 40+ screens
Emotion colour system standardised
Plain-language label audit - all jargon replaced
Consent + privacy layer designed

🧪

VALIDATE

3 moderated usability tests (think-aloud)
ISO 9241-11 usability framework applied
Task completion measured on 5 key flows
SUS scoring across all 3 participants
Morphcast stakeholder review + iteration
WCAG 2.1 AA accessibility audit

4 Core Pain Points Identified

Interface Too Hard to Read

Everything competes for attention, so nothing gets it. No hierarchy, no story, no clear starting point. Users described it as "overwhelming" and "data-heavy".

No Clear Next Steps

Platform tells users what is happening - but never what to do about it. Users want "If students are losing focus, tell me to ask a question."

Unknown Tracking State

Every participant raised concerns about not knowing when facial tracking was active. Consent buried in legal text - not in the product.

Not Accessible

Colour-coded bars with no text labels, small fonts on dark backgrounds, no keyboard nav, broken help pages. WCAG 2.1 AA - not met anywhere.

Professionals Interviewed

Educator · Clinician · Researcher

Screens Designed

40+

Lo-fi + Hi-fi, full system

Usability Rating

3–4/5

All 3 participants post-test

>85%

Task Completion Target

Key flow completion target set and validated through 3 moderated usability sessions with think-aloud protocol.

SUS >75

Usability Score Goal

Target above industry average. Participants described the redesign as "clean, modern, and intuitive" - all rated 3–4/5 for future use.

WCAG AA

Accessibility Compliance

4.5:1 contrast ratios, text labels on every emotion bar, full keyboard navigation, ARIA labels - built in from the start.

>90%

Consent Opt-In Target

Persistent in-session tracking indicator + plain-language consent screen designed to build trust and transparency with all participants.

विस्तार से देखें

Full research, wireframes & usability testing - password protected

काम / WORK

02 · Wellness Tech · AI · Mobile App

PRAAN
AI

Breathe Better. Practice Smarter.

An AI-powered breathwork and yoga coaching app delivering real-time, adaptive breath guidance - closing the feedback loop that no existing app addresses.

RoleEntrepreneurial Lead & UX

ContextStartup / Rutgers

Team3 Members

Interviews21 Conducted

User Interviews
April 2026

15/21

Don't Know If
They Breathe Correctly

$107B

Global Yoga
Market Size

FIRST

Adaptive Breath
Coaching Platform

The Feedback Gap / प्रतिक्रिया की कमी

PROBLEM &
SOLUTION

समस्या और समाधान

THE GAP

"I'm halfway through my inhale and the instructor is already saying exhale."

Interview participant · April 2026

No app listens to your breath

❌ Fixed-pace timers ignore real breath cycles
❌ 15/21 users don't know if they breathe correctly
❌ No safety features for solo breathing conditions
❌ Yoga apps ignore breath entirely
❌ 11/21 tried an app and churned

PRAANAI

"Finally know if you're breathing right."

PraanAI Core Value Proposition

Adaptive breath coaching

✅ Microphone-based real-time breath sensing
✅ Advances only when your breath is complete
✅ Safety features for asthma, anxiety, post-COVID
✅ Yoga + breathwork unified in one session
✅ No hardware - works on any phone or laptop

Mobile App / मोबाइल ऐप

THE APP

तैयार उत्पाद

🎙 Real-Time Breath Sensing

Microphone detects inhale/exhale patterns. AI cues adapt to where you actually are in your breath - not a metronome.

🧘 Guided Practice Chapters

Structured yoga + breathwork sessions with chapter-by-chapter progression and streak tracking.

📊 Breath Quality Dashboard

Abdominal breathing %, rhythm consistency, depth score - session-over-session progress tracking.

🛡 Safety Mode

Detects breath-holding, hyperventilation, unsafe pacing. Designed for asthma, anxiety & post-COVID practitioners.

Design System / डिज़ाइन सिस्टम

PRAAN AI
DESIGN LANGUAGE

रंग · टाइपोग्राफी · घटक

01 / COLORS

#F5F1E8

Warm Parchment - BG

#0A0A09

Near Black - Sessions

#3D7A5A

Forest Green - Brand

#8BAF8F

Sage - Progress

#C5D5C8

Mist - Cards, Dividers

#EF4444

Alert Red - Safety

#6B8EC4

Dusk Blue - Sessions

02 / TYPOGRAPHY

Praan AI

Georgia Italic · Display wordmark

good morning, Ananya

Georgia Regular · H1 greeting

SESSION COMPLETE · 87

System UI 700 · All-caps labels

AI-guided breathwork coaching that adapts to your real breath in real-time.

System UI 400 · Body text

03 / COMPONENTS

BEGIN PRACTICE →

Start plan →

Structural Vinyasa

12 sessions · ★ 4.9

›

FEATURED PLAN

the eight limbs, walked

For you · Yoga · Meditation

04 / SPACING & GRID

4px base

8px gap

16px section

24px padding

Border Radii

4px

8px

14px

pill

Iconography

breath

profile

progress

sleep

Interactive Prototype / इंटरैक्टिव प्रोटोटाइप

THE FULL
PROTOTYPE

पूरा प्रोटोटाइप

🎬 ALL 19 SCREENS - CLICK TO NAVIGATE

iPhone · 9│iPad Pro · 5│MacBook Pro · 5

iPhone · Splash

iPhone · Camera Setup

iPhone · Intent

iPhone · Browse

iPhone · Home

iPhone · Circle

iPhone · Session End

iPhone · Profile

iPhone · Settings

iPad · Landing

iPad · Progress

iPad · Live Session

iPad · Circle

iPad · Home

Mac · Home

Mac · Library

Mac · Circle

Mac · Progress

Mac · Live Session

iPhone · Splash

1 / 19

Design Process / प्रक्रिया

HOW THE
DISH WAS COOKED

कैसे बना यह

🔍

DISCOVER

21 in-depth interviews
7 in-person · 9 Zoom · 5 phone
Competitive analysis: Calm, Headspace, Breathwrk, Wim Hof
RWW Framework applied
Concept screening matrix

🗺️

DEFINE

4 user archetypes mapped
MoSCoW prioritisation
Needs hierarchy established
Target specs + metrics
Hardware pivot to software

✏️

DESIGN

Figma prototype - Figma
Onboarding + session flows
Real-time cue UI patterns
Progress dashboard design
Service blueprint + ecosystem map

🧪

VALIDATE

5-participant concept test
100% understood breath score
Adaptive pacing - top feature
NJII / Y Combinator targets
Patent filing roadmap

Key Interview Signals

The Feedback Gap

15/21 don't know if abdominal breathing is correct. The problem isn't content scarcity - it's the absence of any responsive confirmation loop.

Starting Is Hardest

17/21 cite inertia as the biggest barrier. Not knowing where to begin at home is the primary drop-off point before any session starts.

Pacing Mismatch

"I'm halfway through my inhale and the instructor says exhale." Fixed timers ignore where the practitioner actually is in their breath cycle.

Health-Driven Users

7/21 started yoga due to breathing conditions. 9/21 for anxiety. These users need safety and correctness confirmation - not just content.

User Interviews · April 2026

4.75

Concept Score (top-ranked)

100%

Concept test: understood breath score immediately

≥90%

Breath Detection Accuracy

Target accuracy for inhale/exhale detection via on-device microphone ML - core to the product's value.

≤200ms

Adaptive Pacing Response

Cue delivery latency target - must feel real-time, not delayed, to replace fixed-timer competitors.

≥80%

On-Device Processing

Microphone data processed locally - privacy-first architecture for health-sensitive breath audio.

$107B

Market Opportunity

Global yoga market + $2.1B breathwork app segment + $60B+ corporate wellness - scalable B2B + B2C.

विस्तार से देखें

Full research, interviews, concept screening & product strategy - password protected

काम / WORK

03 · Education · UX Research · AI Design

KNIGHT
PLAN

Smart Scheduling for Scarlet Knights

Redesigning Rutgers University's course registration - replacing fragmented, anxiety-inducing WebReg with an AI-powered platform for 50,000+ students.

RoleLead UX Researcher & Designer

Team4 Members

Duration14 Weeks

CourseContextual Inquiry SP25

Stakeholder
Interviews

90%

Task Completion
Rate · SUS 84

50K+

Rutgers Students
Impacted

72%→11%

Error Rate
Reduction

Problem & Solution / समस्या और समाधान

BEFORE &
AFTER

पहले और बाद में

BEFORE - WEBREG

14 STEPS

to register one course

❌ 5+ fragmented platforms (WebReg, CSP, Degree Nav…)
❌ 72% error rate in first session
❌ No real-time conflict detection
❌ No AI guidance
❌ Mobile completely broken

AFTER - KNIGHT PLAN

4 STEPS

same outcome

✅ One unified platform
✅ Error rate dropped to 11%
✅ Auto conflict detection
✅ AI-powered academic advisor
✅ Mobile-first, WCAG AA

Finished Product / तैयार उत्पाद

THE FULL
PROTOTYPE

पूरा प्रोटोटाइप

🖥 11 SCREENS - CLICK TO NAVIGATE

Welcome Dashboard

AI Advisor - Getting Started

AI Advisor - Registration

Course Search

Course Search - Grid View

Quick Search Overlay

Course Planner

Registration Confirmed

Student Profile

Settings

Help & Support

knight-plan.rutgers.edu

Welcome Dashboard

1 / 11

Design System / डिज़ाइन सिस्टम

KNIGHT PLAN
DESIGN LANGUAGE

रंग · टाइपोग्राफी · घटक

01 / COLORS

#CC0033

Scarlet Red - Brand

#F5F5F5

Light Grey - Page BG

#FFFFFF

White - Cards, Panels

#2C2C2C

Charcoal - Sidebar

#E8D5D5

Blush - Hover states

02 / TYPOGRAPHY

Academic Hub

Helvetica Neue 700 · UI Heading

How can I help you today, Scarlet Knight?

System UI 400 · AI Greeting

SPRING 2024 SEMESTER

System UI 600 · Label caps

03 / COMPONENTS

+ Register for Classes

Recommend Courses

Based on your major

Dashboard · Course Search · AI Advisor

04 / SPACING

4px base

8px gap

16px section

Research Process / शोध प्रक्रिया

HOW THE
DISH WAS COOKED

कैसे बना यह

🔍

DISCOVER

6 stakeholder interviews · Affinity diagram · Day-in-the-Life model

🗺️

DEFINE

4 user personas · Sequence model · Cool Drilldown workshop

✏️

DESIGN

Low-fi wireframes → High-fidelity Figma prototype · 11-screen desktop UI

🧪

VALIDATE

8 usability sessions · SUS 84/100 · Think-aloud protocol

90%

Task Completion

5 key tasks across search, plan, register, drop, smart suggest.

84/100

SUS Score

Classified as "Excellent" - above the industry benchmark of 80.

WCAG AA

Accessibility

High contrast, screen reader support, mobile-first responsive.

AI-First

AI Advisor

"I wish this was the real registration tool." - usability test participant

विस्तार से देखें

Full storyboard, research documentation, ideation & usability testing - password protected

काम / WORK

UX RESEARCH · 2024

SLINGSHOT

डिजिटल कार्यस्थल शोध

Comparing web & mobile user experiences across digital workplace platforms - featuring Slingshot by Infragistics.

UX Research Heuristic Evaluation Competitive Analysis Team Lead

THE BRIEF

Compare and evaluate the user experience of Slingshot alongside three leading digital workplace competitors - across web and mobile - to surface actionable improvement opportunities.

MY ROLE

Team Lead · UX Researcher · Heuristic Evaluator · Ideation Lead

CONTEXT

Rutgers MBS Externship · Infragistics · 15-week program · Team of 4 researchers

01 / PROBLEM STATEMENT

Web & Mobile:
A Tale of Two Experiences

Digital workplace tools often deliver inconsistent experiences across platforms. The web version offers rich features while mobile tends to lag behind - creating friction and inefficiencies for users moving between devices. Our research goal: identify these gaps and generate data-backed recommendations for Slingshot.

RESEARCH GOALS

Understand the differences between web and mobile experiences

Evaluate features, functionality, and design consistency

Develop key insights and ideations to improve Slingshot

02 / APPROACH

15-Week Research Sprint

WK 1–2

Each member selects a product to evaluate

WK 3

Develop use cases and personas

WK 4

Define evaluation criteria

WK 5–10

Individual product evaluations

WK 11–13

Collate findings and synthesise insights

WK 14–15

Create report and present to Infragistics

03 / COMPETITIVE ANALYSIS

Four Products.
One Framework.

Our team evaluated Slingshot alongside three leading digital workplace competitors - each member taking one product and applying a consistent set of use cases and heuristics across both web and mobile platforms.

Slingshot

by Infragistics

Competitor 1

Task & Project Mgmt

Competitor 2

Workspace Management

Competitor 3

Work OS Platform

Use Cases Evaluated

Onboarding - signing into the product and first-use experience

First overview - clarity of the initial dashboard

Profile setup - photo, basic info, notification preferences

Creating hierarchy - workspace, project, list structure

Task creation - name, assignees, timeline, priority, subtasks

Task progression - status updates and team visibility

Communication - task comments and team collaboration

Task views - Grid vs Calendar vs Kanban, filtering options

04 / HEURISTIC EVALUATION

Nielsen's 10 Heuristics
Across All Platforms

HEURISTIC	S	C1	C2	C3
1. Visibility of System Status	10	9	8	6
2. Match Between System & Real World	10	7	9	10
3. User Control & Freedom	8	9	9	8
4. Consistency & Standards	9	9	7	7
5. Error Prevention	8	8	4	10
6. Recognition Rather than Recall	4	9	9	7
7. Flexibility & Efficiency of Use	9	8	7	10
8. Aesthetic & Minimalist Design	6	10	9	8
9. Help Users Recognise & Recover from Errors	10	9	2	10
10. Help & Documentation	8	10	10	8

S = Slingshot · C1 = Competitor 1 · C2 = Competitor 2 · C3 = Competitor 3 · Scored /10

05 / SLINGSHOT FINDINGS

Web vs. Mobile:
The Gap Analysis

WEB PLATFORM

STRENGTHS

Advanced versatile task-handling
Intuitive visual design
Deep customisation options
Dashboard analytics from datasets

WEAKNESSES

No onboarding tutorial
Missing undo/redo functionality
Minor formatting/crash issues
Inefficient team assignment UX

OPPORTUNITIES

Toggled onboarding tutorial
Universal create + undo buttons
AI-powered templates
Personalised dashboard views

MOBILE PLATFORM

STRENGTHS

Feature-rich for mobile
Real-time notifications
Easy collaboration channels

WEAKNESSES

Cluttered layout on small screens
Web-dependent for key actions
Compact calendar - poor visibility
Cannot check off tasks inline

OPPORTUNITIES

Mobile-specific gestures
Simplified web features for mobile
Expanded calendar with task list
AI chat + workspace templates

06 / IDEATION

4 Key Design
Recommendations

Redesign Onboarding

Offer users a choice: tailor their experience by setting up their profile, or jump straight into the product. Speed matters - reduce friction at first contact.

Improve Tutorial UX

Add close and continue buttons. Make imagery larger and more contextual. Support swipe and drag gestures on mobile for a fluid learning experience.

AI Integration

Add an AI chatbot and AI-generated workspace templates on mobile. Introduce a universal "Add" button in the main navigation for quick task, workspace, and chat creation.

Redesign Calendar View

Make the mobile calendar more intuitive - larger task indicators, swipe navigation between months, and a scrollable task list beneath the calendar view.

07 / KEY LEARNINGS

🔬

Research Depth

This project expanded my toolkit - heuristic evaluation, competitive benchmarking, and ideation within real product constraints.

📱

Platform Parity Matters

Consistent features across web and mobile are not a luxury - they're essential. The gap between platforms directly impacts how much users trust and rely on a product.

🤝

Collaborative Research

Every team member brought a unique lens. Synthesising four separate evaluations into unified insights taught me that great research is always a team sport.

आ ला कार्ट / À LA CARTE

À LA CARTE · 3D

3D PRODUCT
DESIGN

त्रि-आयामी उत्पाद

3D product visualisations exploring form, material, and function - from everyday tools to consumer electronics. Each render focuses on accurate geometry, lighting, and surface quality.

PRODUCT RENDER

Electric Guitar

PRODUCT RENDER

Bosch Hand Drill

PRODUCT RENDER

LG Garment Styler

PRODUCT RENDER

Hair Straightener

आ ला कार्ट / À LA CARTE

À LA CARTE · ANIMATION

3D ANIMATION

3डी एनिमेशन

Motion and product animation - breathing life into 3D objects through movement, physics, and storytelling.

3D ANIMATION

Tampon Dispenser

3D ANIMATION

Drone Render

आ ला कार्ट / À LA CARTE

À LA CARTE · EXPERIMENTS

PLAYGROUND

प्रयोगशाला

A space for creative experiments - side projects, AI collaborations, and design play that lives outside the brief.

AI EXPERIMENT

AI Drawing Book

An experiment in AI-assisted illustration - exploring how generative tools can extend creative practice into uncharted territories.

Altasparq Aware · Deep Dive

THIS DISH IS
CHEF'S TABLE
ONLY

Full process documentation. Ask Esha for the secret ingredient.

ALTASPARQ AWARE · FULL PROCESS

🔒 Confidential

Full Research & Design Documentation

ALTASPARQ
AWARE

A complete documentation of the UX research and design process behind the redesign of Altasparq Aware - an emotion-AI platform for clinical video sessions. From heuristic evaluation through to validated high-fidelity prototype.

RoleUX Designer

InstitutionRutgers University

Duration14 Weeks

Team4 Members

Interviews5 Conducted

What's Inside

01 · Competitive Analysis 02 · Heuristic Evaluation 03 · SWOT Analysis 04 · User Research 05 · Interviews 06 · Personas 07 · Journey Mapping 08 · Ideation 09 · Wireframes 10 · High-Fidelity Design 11 · Usability Testing 12 · Findings & Outcomes

COMPETITIVE
ANALYSIS

प्रतिस्पर्धी विश्लेषण

We benchmarked Altasparq Aware against two leading emotion AI platforms to understand where it lagged and where it could differentiate. This shaped our core redesign priorities.

Affectiva (Affdex / iMotions)

Mature, widely adopted platform with polished UI. However, it lacks proactive in-session alerts and post-session plain-language summaries. Still relies on colour-coding without accessibility fallbacks.

Navigation ClarityModerate Plain-language LabelsPartial Actionable GuidanceMinimal AccessibilityPartial Onboarding SupportBasic

MyNoldus FaceReader

Research-grade tool for academic psychology labs. Interface is dense, technically complex, requires significant training. No actionable guidance - prioritises precision over clarity.

Navigation ClarityPoor Plain-language LabelsNone Actionable GuidanceNone AccessibilityPoor Onboarding SupportManual

Key Takeaway

Both competitors suffer from the same core problem as Altasparq: they prioritise data output over human understanding. Our opportunity was to be the first platform that genuinely closes the gap from data to action - for users without a clinical or technical background.

DimensionAffectivaFaceReaderAltasparq NowAltasparq Target

NavigationModeratePoorPoorStrong

Plain LanguagePartialNoNoThroughout

Actionable GuidanceMinimalNoneNoneEvery view

AccessibilityPartialPoorPoorWCAG AA

OnboardingBasicManualBrokenInteractive

HEURISTIC
EVALUATION

अनुमानी मूल्यांकन

We applied Nielsen's 10 Usability Heuristics across every major screen of the current Altasparq platform. Each violation was rated for severity (1–4) and frequency.

Severity 4 · Catastrophic

Visibility of System Status

No persistent indicator that facial tracking is active. Users have no way to know if analysis is running during a session.

Severity 4 · Catastrophic

Help & Documentation

FAQ and Support pages return "Page not found." New users have zero self-service path. Broken from day one.

Severity 3 · Major

Match Between System & World

Labels like "FER Stream", "Valence", and "Arousal" mean nothing to educators, therapists, or corporate trainers.

Severity 3 · Major

Aesthetic & Minimalist Design

Dashboard shows all data simultaneously with equal visual weight - no hierarchy, no focal point. Everything competes for attention.

Severity 3 · Major

Consistency & Standards

Mixed chart types (bar, gauge, line) in the same view. Same emotion can appear in different shades across screens. Icons not standardised.

Severity 2 · Minor

User Control & Freedom

No pause toggle visible during sessions. Stop Analysis button exists but is not prominent. Users feel locked in once tracking starts.

Severity 3 · Major

Error Prevention

If camera is blocked or face moves offscreen, platform shows zeroes with no error message, guidance, or recovery path.

Severity 3 · Major

Recognition over Recall

Navigation requires memorisation. No breadcrumbs on any subpage. Three separate contact links (Support, Feedback, Report Abuse) with no clear distinction.

Evaluation Summary

12 distinct heuristic violations identified. 4 rated Severity 4 (catastrophic), 6 rated Severity 3 (major). No core screen passed all 10 heuristics. WCAG 2.1 AA compliance was not met on any screen evaluated.

SWOT
ANALYSIS

SWOT विश्लेषण

A strategic assessment of where Altasparq stands - what gives it an advantage, what's holding it back, and where the real opportunity lies.

Strengths

Morphcast AI detects 30+ emotional states in real time - genuinely powerful technology
Proactive in-session alerts are a genuine differentiator vs. all competitors
Cloud-based - no specialist hardware required
Scenario templates offer a configurable starting point for different user types

Weaknesses

Navigation fragmented and inconsistently labelled across screens
Visual design is dense, dark, and near-impossible to scan quickly
No plain-language summaries or actionable guidance for non-specialists
Onboarding pages broken - FAQ returns 404
Accessibility does not meet WCAG 2.1 standards on any screen

Opportunities

A cleaner redesign directly removes the primary barrier to adoption
Modular dashboard can serve diverse users (education, healthcare, corporate)
Strong onboarding reduces churn among new users
Accessible design opens the platform to significantly more users
Plain-language summaries create a new "non-specialist" user segment

Threats

Competitors may improve UX before Altasparq acts on research
Privacy concerns (GDPR, HIPAA) require transparent design - current state doesn't meet this bar
Without trust-building redesign, adoption will continue to stall
Regulatory pressure on facial recognition is increasing globally

USER
RESEARCH

उपयोगकर्ता शोध

Our research focused on professionals who actually use or would use emotion-AI in their work. We combined in-depth interviews with a literature review and competitive benchmarking to triangulate findings.

In-depth Interviews

Professional Domains

Heuristic Violations Found

Core Pain Points Mapped

Research Goals

→Understand how real professionals interact with emotion-AI data during live sessions
→Identify specific pain points that prevent users from extracting value from the platform
→Understand trust and privacy concerns around facial tracking in professional contexts
→Surface latent needs - things users want but haven't been able to articulate

Research Methods

01Semi-structured in-depth interviews (think-aloud protocol)
02Heuristic evaluation using Nielsen's 10 heuristics
03Competitive analysis (Affectiva, MyNoldus FaceReader)
04Affective computing literature review
05Affinity mapping & thematic analysis of findings

USER
INTERVIEWS

उपयोगकर्ता साक्षात्कार

5 in-depth interviews with professionals across education, healthcare, and research. Each session used a think-aloud protocol and lasted 45–60 minutes. Participants were shown the live platform and asked to complete key tasks.

👩‍🏫

Dr. Anita Deshmukh

High School Science Teacher · Edison, NJ · 14 years experience

Cognitive Overload Jargon Confusion Actionable Guidance

"The dashboard shows too many data points at once - I need one clear signal, not fifteen. And if I could see that students are losing focus, tell me to ask a question - don't just show me a percentage."

→Needs one-glance engagement signal, not multi-panel data

→"FER Stream" and "Arousal" are meaningless without explanation

→Wants to know when tracking is active - currently invisible

→Google Meet and Canvas LMS integration is non-negotiable

👩‍🎓

Neha Patel

Psychology Graduate Student · Jersey City, NJ · Teaching Assistant

Visual Hierarchy Privacy Concerns Plain Language

"The layout has no clear visual hierarchy - everything feels equally important, so nothing stands out. Tell me what happened in a sentence, then I'll look at the charts if I want more detail."

→Technical terms need plain-language equivalents or inline tooltips

→Post-session data should lead with a summary sentence, not charts

→No visible indicator of when facial tracking is running - discomfort

→Background in cognition - comfortable with concept, not with UI

👩‍⚕️

Brooke Shaw

Speech-Language Pathologist · New York, NY · 1.5 years clinical experience

Screen Fatigue Workflow Fit Customisation

"The in-session dashboard is too large - it takes up too much screen space during patient interactions. I need a picture-in-picture option, not a full split-screen. Dark mode by default is also a problem for clinical settings."

→Wants compact, toggleable PiP panel - not a full sidebar

→Real-time graph animation is visually fatiguing over long sessions

→Needs light mode option with strong contrast for clinical environments

→Wants saved dashboard configs per patient - no re-setup per session

Unified Themes Across All Interviews

→

Cognitive Overload - "Overwhelming", "data-heavy", "too much at once"

→

Jargon & Clarity - "FER Stream means nothing to me"

→

No Actionable Guidance - "Shows data but doesn't say what to do"

→

Privacy Uncertainty - "I don't know when tracking is active"

PERSONAS

उपयोगकर्ता व्यक्तित्व

We built three research-grounded personas directly from interview findings - not aspirational demographics. Each represents a distinct set of goals, frustrations, and design requirements.

👩‍🏫

Priya - The Educator

High School Teacher, 38, Edison NJ

"I need to know what's happening at a glance - not decode a second screen while I'm teaching."

Goals

Monitor student engagement without distraction
One-click access to what's happening right now
LMS integration with Google Meet & Canvas

Frustrations

Too many panels - can't process while teaching
Technical jargon with no explanation
No actionable suggestions from the data

Design Need: Single engagement summary card + disengagement alerts

👩‍⚕️

Brooke - The Clinician

Speech-Language Pathologist, 29, New York NY

"The dashboard takes up too much space. I need it compact, toggleable, and out of my way during patient sessions."

Goals

Minimal screen footprint during live patient sessions
Save configurations per patient - no re-setup
Light mode + strong contrast for clinical settings

Frustrations

Full split-screen is too large for clinical use
Real-time graph animation causes visual fatigue
Dark mode inappropriate for clinical environments

Design Need: Compact PiP panel · per-patient saved layouts

🔬

Neha - The Researcher

Psychology Graduate Student, 25, Jersey City NJ

"I understand what the platform does - I just can't understand what it's trying to show me."

Goals

Access post-session summaries in plain language
Understand emotional trends across multiple sessions
Export data for academic research

Frustrations

No plain-language alternative to technical charts
No visible consent indicator during tracking
No data export option anywhere in interface

Design Need: Plain-language summaries · data export · consent layer

JOURNEY
MAPPING

यात्रा मानचित्रण

We mapped the full experience for each persona across five stages. Here we show Priya's journey - the educator - before and after the redesign, at each touchpoint.

StageGoalActionsPain Point (Current)Redesign SolutionEmotion

Awareness Learn about platform Reads about Altasparq online Sceptical about privacy Clear messaging on transparency + consent 😐 Curious

Onboarding Set up account Creates account; starts demo session "FER Stream" labels confusing; FAQ broken (404) Guided tooltips; plain-language throughout; working Quick Start 😟 Confused

In-Class Use Monitor live engagement Glances at dashboard while teaching Too many graphs to process simultaneously Single insight card; disengagement alerts 😤 Frustrated

Post-Session Reflect on what happened Reviews session summary No clear actions or next steps "What to try next time" suggestion section 😕 Underwhelmed

Long-Term Evaluate continued use Shares feedback with colleagues Unsure about data use and storage Transparent data panel; clear consent record 😊 Cautiously Positive

IDEATION &
CONCEPTS

विचार मंथन

With research findings mapped to design requirements, we ran two ideation sessions with the full team. We used "How Might We" questions as starting points, then sketched rapid concepts and voted on which directions to prototype.

How Might We Questions

How might we give users one clear signal instead of fifteen metrics - without hiding capability?

How might we make the platform feel like it's working with the user rather than just reporting at them?

How might we build trust and transparency around facial tracking directly into the product experience?

How might we serve an educator, a therapist, and a researcher with the same core system but different views?

Top 3 Concept Directions (from dot voting)

Progressive Disclosure Dashboard

Show one primary insight card front and centre. All graphs and detail panels accessible on demand - never forced. Modular layout per user role.

🗳 9/10 votes · Carried forward

Recommendations Engine

Every data insight is paired with a suggested action. "Engagement dropping - try asking an open question." Transforms platform from monitor to assistant.

🗳 8/10 votes · Carried forward

Persistent Consent Layer

Before-session consent screen. Persistent "Analysis Active" banner during. Clear pause toggle always visible. Data transparency panel in settings.

🗳 10/10 votes · Non-negotiable

WIREFRAMES

वायरफ्रेम

We built low-fidelity wireframes to test core structural decisions before investing in visual design. Two dashboard layout alternatives were A/B tested with 5 participants. Alternative B (flat cards, consistent spacing) won.

Alternative A · Drop Shadows

❌ Not selected - visually complex, heavier cognitive load

Alternative B · Flat Cards ✓ Selected

✅ Selected - clean, minimal, faster to scan

Key Low-Fi Decisions

Navigation Consolidation

Support, Feedback, Report Abuse merged into single "Help" page. Active-page indicator changed from thin underline to reverse-colour highlight.

Modular Dashboard

Drag-and-drop components. Pre-set templates for Education, Healthcare, Business so new users aren't faced with a blank canvas.

In-Meeting Sidebar

Persistent vertical sidebar replaces full-screen takeover. Notification panel shows real-time alerts as short cards, not continuous graphs.

Onboarding Checklist

"Start Here" popup for first-time users. Minimisable at any point. "Get Started" tab always one click away.

HIGH-FIDELITY
DESIGN

उच्च-विश्वसनीयता डिज़ाइन

Following stakeholder feedback from Morphcast on the lo-fi prototype, we moved into high-fidelity design. All structural decisions were carried forward. Focus shifted to visual refinement, interaction detail, and full design system consistency.

🏗

Configurable Dashboard Hub

Central hub with drag-and-drop components, role-appropriate templates, and a clearly labelled "Customise" button. Edit mode visually highlights what can be moved.

🎥

In-Meeting Interface

Main Insights Box always visible with one clear current message. Tracking indicator persistent. Pause/Resume one click away. All emotion bars include text labels - colour is never the only signal.

📋

Post-Session Summary

Top 3 emotions, engagement score, attention trend, and "What to try next time" - all in plain language. Designed to be understood in under 30 seconds. PDF/CSV export in one click.

♿

Accessibility Settings

Light, Dark, Flat themes with live previews. Typography controls (Sans-serif, Serif). Text resize, high contrast, dyslexia-friendly font. ARIA labels on every interactive element.

🔒

Consent & Transparency Layer

Plain-language consent screen before every session. Persistent "Emotion Analysis Active" banner during. Pause toggle always visible. Data retention explained in plain English in Settings.

🎓

Contextual Onboarding

Pop-up instructions on any screen with complex features. Tooltips on hover for every technical element. Quick Start guide uses images, short animations - not walls of text.

USABILITY
TESTING

उपयोगिता परीक्षण

3 moderated usability testing sessions on the high-fidelity prototype. Think-aloud protocol. Participants explored Dashboard, Meeting Page, Accessibility Settings, and Onboarding. Tasks included adjusting dashboard layout, reviewing session insights, switching themes.

Participant 1 · Gary

★★★★4 / 5 likely to use

"Clean and modern. Layout feels familiar. Customisation controls could be more discoverable - the icon in the top-right was easy to miss."

✅ Dashboard components intuitive

✅ Navigation straightforward

⚠ Customise icon not discoverable

⚠ Editable areas need visual highlight

Participant 2 · Joel

★★★★4 / 5 likely to use

"Clean and organised, a little dense in some areas. Drag-and-drop in edit mode felt intuitive. Dark mode toggle was easy to find."

✅ Edit mode drag-and-drop natural

✅ Theme previews helpful

⚠ Placeholder boxes need more contrast

⚠ Requested mute/spotlight in emotion panel

Participant 3 · Anonymous

★★★½3–4 / 5 likely to use

"The modular concept is strong. I navigated to edit mode correctly. I'd love a line graph showing how emotions changed over the session."

✅ Edit interface "clear and straightforward"

✅ Modular concept well received

⚠ Requested emotional timeline chart

⚠ Dashboard views feel disconnected

What Worked vs. What to Improve

Tested Well

✅ Clean layout; familiar structure; intuitive components

✅ Drag-and-drop edit mode felt natural

✅ Dark mode and typography controls well-received

✅ Sidebar navigation clear and consistent

✅ Quick Start guide clear and helpful

Needs Iteration

⚠ Customise button needs clearer label alongside icon

⚠ Editable areas in edit mode need visual highlight

⚠ Emotional timeline chart missing from meeting view

⚠ Page-to-page flow needs breadcrumbs

⚠ Advanced settings should move to sub-menu

FINDINGS &
OUTCOMES

निष्कर्ष और परिणाम

The redesign was validated across three usability sessions. All participants described the interface as a significant improvement. Core design decisions - modular dashboard, plain-language labels, single-insight summary, visible consent - all tested positively.

>85%

Task Completion Target

On all 5 key user flows tested

SUS >75

Usability Score Goal

Above industry average benchmark

WCAG AA

Accessibility

4.5:1 contrast, full keyboard nav, ARIA labels

3–4/5

Likelihood to Use

All 3 usability test participants

>90%

Consent Opt-in Target

Transparent in-session tracking layer

NPS >40

Net Promoter Score

Target for post-launch measurement

The Bottom Line

"The technology is not the barrier to adoption. The design is. Fix the design, and the product works." - Consolidated testing finding.

End of Process Documentation

Altasparq Aware · Esha More · Rutgers University 2025

PraanAI · Full Process

THIS RECIPE IS
BREATHWORK ONLY

Full product research & process documentation. Ask Esha for the password.

PRAANAI · FULL PROCESS DOCUMENTATION

🔒 Confidential

Full Research & Product Strategy Documentation

PRAAN
AI

Complete documentation of the entrepreneurial research, product strategy, and UX design process behind PraanAI - from initial smart mirror concept through validated AI breathwork app.

RoleEntrepreneurial Lead & UX

TeamEsha More · Nandini · Vishnu Iyengar

Interviews21 Conducted · April 2026

What's Inside

01 · Opportunity Identification 02 · Competitive Analysis 03 · User Interviews (21) 04 · User Archetypes 05 · Target Specifications 06 · Product Architecture 07 · Prototype & Testing 08 · Findings & Roadmap

OPPORTUNITY
IDENTIFICATION

अवसर की पहचान

The founding team generated 5 product concepts before converging on PraanAI. Each idea was screened against user pain, market opportunity, and technical feasibility.

✓ Selected - AI Breathwork App

Mobile/web app using microphone-based breath sensing and adaptive AI coaching. No hardware required. Works on any phone or laptop. Directly addresses the #1 pain: not knowing if you're breathing correctly.

Concept Score: 4.75 / 5.00 · Ranked #1

✗ Eliminated - Smart Mirror

Original concept: AI-powered smart mirror with depth sensors for yoga pose correction. Eliminated after interviews revealed practitioners actively keep devices off the mat. Hardware creates friction before the session starts.

Concept Score: 2.75 / 5.00 · Ranked #4

→ Phase 2 - Instructor Dashboard

B2B tool for yoga instructors to monitor student breath quality between sessions. Assign sessions, review progress reports, target correction in next class. Strong signal from instructors in interviews.

Concept Score: 3.90 / 5.00 · Phase 2

✗ Eliminated - Other Concepts

Bus Stop Wellness Module (high infrastructure cost), AI Fridge Assistant (outside wellness domain), and Wearable Pendant (merged as safety feature layer). All eliminated based on feasibility and user mismatch.

Eliminated in screening

Key Pivot Insight

Our initial concept centred on hardware (smart mirror) because hardware felt premium. User interviews corrected this assumption quickly: practitioners do not want a fixed device on the mat. The pivot from hardware to software unlocked the core value proposition - a breath-responsive coach that works on any device.

COMPETITIVE
ANALYSIS

प्रतिस्पर्धी विश्लेषण

We benchmarked PraanAI against all major breathwork and yoga platforms. The finding was consistent: no competitor listens to and responds to the user's actual breath.

PlatformBreath SensingAdaptive PacingYoga IntegrationHealth Focus

CalmNoneFixed timerNoPartial

HeadspaceNoneFixed timerPartialPartial

BreathwrkNoneFixed timerNoPartial

Wim HofNoneFixed timerNoPartial

Down DogNoneNoneYesNo

PraanAI (Target)≥90% accuracyBreath-responsiveYesPrimary focus

The Gap

No major breathwork app delivers genuine real-time breath-responsive feedback. PraanAI's differentiation: breath-audio sensing + adaptive cue delivery - a product category that does not yet exist.

USER INTERVIEWS
(21 CONDUCTED)

उपयोगकर्ता साक्षात्कार

21 in-depth interviews in April 2026 - 7 in-person at Broome Street Yoga NYC, 9 via Zoom, 5 by phone. Participants: yoga practitioners (beginner to advanced), people with breathing health issues, home-only practitioners.

Total Participants

17/21

Say Starting Is Hardest

15/21

Don't Know If Breathing Right

13/21

Use YouTube as Primary Tool

11/21

Tried an App & Churned

7/21

Started Due to Breathing Conditions

🧘

Home Practitioner Archetype

Health-motivated · Practices alone · No instructor access

"I don't breathe properly. My doctor pointed it out - really shallow, only using the top of my chest."

→Wants confirmation that technique is correct, not just content

→Would pay for a tool that closes the feedback loop their doctor opened

→Doesn't need more videos - needs a responsive coach

😤

Pacing-Frustrated Practitioner

Intermediate · Uses YouTube · Frustrated with fixed timers

"I'm halfway through my inhale and the instructor is already saying exhale."

→Fixed-timer apps create more frustration than guidance

→Adaptive pacing described as "the thing I've always wanted"

→Would switch from YouTube for responsive breath coaching

💭

Uncertain Practitioner

Regular practitioner · No formal training · Self-doubt about technique

"I still don't know if my abdominal breathing is correct. I just know I feel better."

→Felt improvement but no confirmation - anxious about bad habits

→Progress tracking with a breath quality score would be transformative

→One simple score (up/down) preferred over complex dashboards

USER
ARCHETYPES

उपयोगकर्ता प्रकार

Four distinct user archetypes emerged from the 21 interviews - each with a distinct pain and PraanAI value proposition.

🏥

The Pain-Driven Pragmatist

~6/21 · Referred by doctor · Breathing condition

"My doctor told me to do breathwork. But I have no idea if I'm doing it safely."

Primary Pain

Needs safe, guided solo practice
No tool accounts for asthma/anxiety
Doctor referral has no specific app

PraanAI Value: Breath sensing + safety cues for solo practice

😟

The Stressed Beginner

~7/21 · Started for anxiety · Can't stay consistent

"I start a streak and then miss two days and feel like I've failed."

Primary Pain

Inertia - starting is the hardest part
No accountability at home
Content fatigue within weeks

PraanAI Value: Adaptive pacing + streak + accountability

🏆

The Committed Home Practitioner

~5/21 · Daily practice · Never validated technique

"I feel better but I've never had anyone confirm I'm actually doing it right."

Primary Pain

No technique confirmation loop
Uncertain about abdominal breathing
No session-over-session progress data

PraanAI Value: Real-time feedback + progress tracking

📊

The Serious Aspirant

~3/21 · Advancing technique · Wants data

"I want to see my abdominal breathing percentage improve week over week."

Primary Pain

No breath quality metrics anywhere
No instructor-comparable benchmarks
No data-driven improvement path

PraanAI Value: Breath analytics + instructor benchmarks

TARGET
SPECIFICATIONS

लक्ष्य विनिर्देश

Concrete, measurable targets derived from interview frequency and health impact - each tied to a specific user need.

Customer NeedMetricTargetRationale

Real-time breath coachingDetection accuracy≥ 90%Core feature reliability

Adaptive pacingAdjustment latency≤ 200msMust feel real-time

Progress trackingBreath metrics tracked≥ 5 parametersDepth, rhythm, abdominal %, duration, consistency

Mobile deliveryPlatform supportiOS + Android + WebDevice-agnostic; no hardware

Safety featuresAnomaly detection rate≥ 85%Protects breathing condition users

PrivacyOn-device processing≥ 80%Microphone data local - user trust

AffordabilityConsumer subscription≤ $15/monthBelow major wellness app benchmarks

PRODUCT
ARCHITECTURE

उत्पाद वास्तुकला

PraanAI is organised into 6 functional layers, from user-facing interfaces through to deployment infrastructure.

User Interface & Interaction

Mobile app (iOS/Android), Progressive Web App, real-time breath cue overlays, posture alignment guides, session player.

Sensing & Breath Detection

Microphone-based audio sensing, on-device ML for inhale/exhale classification, environmental noise filtering, Phase 2 camera posture overlay.

Intelligence & AI Processing

Breath Analysis Engine, Adaptive Coaching Engine, Progress & Analytics Engine, Safety Monitor for breath anomalies.

Data & Integration

Cloud database, RESTful APIs, end-to-end encryption, ≥80% on-device processing, Phase 2 B2B instructor API.

Motivation & Engagement

Progress visualisation, streak tracking, reminders, Phase 2 peer accountability circles and instructor-assigned sessions.

Deployment & Scalability

AWS/GCP cloud with renewable energy, edge-first architecture, cross-platform backend, GDPR/HIPAA-aware data handling.

PROTOTYPE
& TESTING

प्रोटोटाइप और परीक्षण

The Figma prototype validated the user experience of PraanAI's core flows. Concept testing with 5 participants confirmed the value proposition before engineering investment.

Prototype Screens

Onboarding

Brand intro "Breathe Better. Practice Smarter." Health intake: breathing conditions, experience level, primary goal.

Home Screen

Practice streak, recommended session, quick access to breath-focused or yoga+breath session types.

Active Session

Real-time breath cue overlays, adaptive pacing indicator, live breath quality meter updating as session progresses.

Progress Dashboard

Breath Quality Score, Abdominal Breathing %, Rhythm Consistency, session comparison, improvement tips.

Concept Test Results (5 participants)

100% Understood Breath Score

"Finally answers whether I'm doing it right."

✅ Core value immediately grasped

100% Preferred Adaptive Pacing

"The thing I've always wanted from a breathwork app."

✅ Top differentiator confirmed

Health Framing Resonated Deeply

"Practice safely when alone. Know your technique is correct."

✅ Strongest purchase intent trigger

Simple Score Preferred

"One prominent score with trend direction - not a complex dashboard."

⚠ Keep metrics minimal and scannable

FINDINGS &
ROADMAP

निष्कर्ष और रोडमैप

PraanAI is positioned to be the first truly adaptive breathwork coaching platform. The roadmap moves from discovery through grant funding to patent filing and MVP launch.

Phase 1

Customer Discovery

21 interviews complete · Concept validated · Product thesis confirmed

Phase 2

Grant Funding + Prototype

NJII · Y Combinator targets · Prototype development post-grant

Phase 3

Patent + MVP Launch

Breath-detection-to-coaching pipeline · Adaptive cue algorithm · MVP launch

Phase 4

B2B Expansion

Studio dashboard · Corporate wellness · Clinical partnerships · Doctor referral pathway

The Core Thesis

"The feedback gap - not content scarcity - is the sharpest pain. PraanAI does not add more yoga content. It closes the loop that every existing app leaves open: does the practitioner know they are breathing correctly, right now?"

End of Process Documentation

PraanAI · Esha More · Rutgers University 2026

Knight Plan · Full Process Documentation

PASSWORD
REQUIRED

Full research process, storyboard, affinity diagrams, and testing documentation for Knight Plan.

Full Process / पूरी प्रक्रिया

KNIGHT PLAN
RESEARCH PROCESS

Smart Scheduling for Scarlet Knights - contextual inquiry, storyboard, ideation, prototyping and usability testing.

Team

Sanyam Bhat · Vishnu Iyengar · Esha More · Vibha Mugwe

Date

May 06, 2025

01 · Storyboard 02 · Research Methods 03 · Key Findings 04 · Ideation 05 · Prototype Testing 06 · Outcomes

STORY
BOARD

कहानी चित्रपट

Comic-style storyboard capturing the emotional journey of a Rutgers student navigating WebReg - from alarm-setting to final registration relief.

"I need to wake up early so I don't miss the registration window"

Sets alarm to not miss registration window. Anxious

"I really need to find the best professors"

Browses Rate My Professors. Overwhelmed

"Ugh, I forgot my password again…"

Struggles to log into WebReg. Frustrated

"Which classes should I take?"

Searches courses - open/waitlist confusion. Confused

"Why is this taking so long?"

Platform loading slowly at peak time. Impatient

"Open, open, open! Why are so many full?"

Many sections already full. Worried

"I just hope it works. Please no errors…"

Submits registration nervously. Hopeful

"Wait, did I pick the right section?"

Uncertain about section choices. Uncertain

"Phew, I did it! My schedule's set!"

Finalized schedule - relief! Relieved ✓

RESEARCH
METHODS

शोध विधियाँ

Contextual inquiry using mixed qualitative methods including interviews, observations, affinity diagrams, and usability testing.

6 Stakeholder Interviews

3 Students · 1 Academic Advisor · 1 Registrar Staff · 1 IT Developer. Semi-structured, 20 min each.

8 Usability Test Participants

4 undergrad + 4 graduate students. In-person and Zoom. 30–45 min with think-aloud protocol.

Affinity Diagram

Interview data organized into themes using Figma AI-assisted pattern recognition.

Contextual Design Models

Day-in-the-Life, Sequence Model, Identity Model, and User Environment Design.

KEY
FINDINGS

मुख्य खोजें

System Fragmentation

Students navigate 5+ platforms with no integration.

No Real-Time Feedback

No seat counts, no conflict detection, system crashes at peak registration.

Lack of Guidance

No onboarding for first-years. 72% error rate in first session.

Demand for AI

Students want intelligent suggestions, conflict resolution, and an AI advisor.

Interview Quotes

"What aspects of WebReg did I find most challenging? Almost everything on my first session - there's no guidance at all."

"If you could make one change to WebReg, what would it be? A chatbot - something that answers my questions instantly."

IDEATION &
COOL DRILLDOWN

विचारधारा

From Cool Drilldown Workshop to wireframes - applying Sensation & Flow, Connection, Identity, and Spirit frameworks.

Sensation & Flow

Clean, calming UI - stark contrast to overwhelming WebReg. Mobile-first.

Connection

Share schedules with advisors. Collaborative scheduling with study groups.

Identity

Degree audit integration empowers students to own their academic path.

Spirit & Transformation

Multi-semester planning transforms Knight Plan into a graduation partner.

PROTOTYPE
TESTING

प्रोटोटाइप परीक्षण

90%

Task Completion Rate

SUS Score "Excellent"

2.1m

Avg Time Per Task

What Users Said

"It's so intuitive. I just dropped in my courses like a puzzle!"

"I wish this was the real registration tool."

OUTCOMES

परिणाम

72%→11%

Error Rate Reduction

From 72% error rate in old WebReg to 11% after redesign.

14→4

Steps to Register

From 14 fragmented steps across platforms to 4 unified steps.

WCAG AA

Accessibility

Full mobile-first responsive with screen reader support.

50K+

Students Served

Designed to serve all Rutgers undergrad and graduate students.

TODAY'SSPECIALS

ALTASPARQAWARE

PRAANAI

3 projects3 industries1 obsession:

KNIGHTPLAN

THIS CHEFWEARS MANY HATS

WHY THIS.WHY ME.

EVERYTHINGON THE TABLE

WHAT OUR GUESTSARE SAYING

MAKE ARESERVATION

ALTASPARQAWARE

BEFORE &AFTER

The Problem

The Solution

THE FINISHEDPRODUCT

ALTASPARQ DESIGN SYSTEM

HOW THEDISH WAS COOKED

PRAANAI

PROBLEM &SOLUTION

No app listens to your breath

Adaptive breath coaching

THE APP

PRAAN AIDESIGN LANGUAGE

THE FULLPROTOTYPE

HOW THEDISH WAS COOKED

KNIGHTPLAN

BEFORE &AFTER

THE FULLPROTOTYPE

KNIGHT PLANDESIGN LANGUAGE

HOW THEDISH WAS COOKED

SLINGSHOT

Web & Mobile:A Tale of Two Experiences

15-Week Research Sprint

Four Products.One Framework.

Nielsen's 10 HeuristicsAcross All Platforms

Web vs. Mobile:The Gap Analysis

4 Key DesignRecommendations

Redesign Onboarding

Improve Tutorial UX

AI Integration

Redesign Calendar View

Research Depth

Platform Parity Matters

Collaborative Research

3D PRODUCTDESIGN

Electric Guitar

Bosch Hand Drill

LG Garment Styler

Hair Straightener

3D ANIMATION

Tampon Dispenser

Drone Render

PLAYGROUND

AI Drawing Book

THIS DISH ISCHEF'S TABLEONLY

ALTASPARQAWARE

COMPETITIVEANALYSIS

HEURISTICEVALUATION

SWOTANALYSIS

USERRESEARCH

USERINTERVIEWS

PERSONAS

JOURNEYMAPPING

IDEATION &CONCEPTS

WIREFRAMES

HIGH-FIDELITYDESIGN

USABILITYTESTING

FINDINGS &OUTCOMES

THIS RECIPE ISBREATHWORK ONLY

PRAANAI

OPPORTUNITYIDENTIFICATION

COMPETITIVEANALYSIS

USER INTERVIEWS(21 CONDUCTED)

USERARCHETYPES

TARGETSPECIFICATIONS

PRODUCTARCHITECTURE

PROTOTYPE& TESTING

FINDINGS &ROADMAP

KNIGHT PLANRESEARCH PROCESS

STORYBOARD

TODAY'S
SPECIALS

ALTASPARQ
AWARE

PRAAN
AI

3 projects
3 industries
1 obsession:

KNIGHT
PLAN

THIS CHEF
WEARS MANY HATS

WHY THIS.
WHY ME.

EVERYTHING
ON THE TABLE

WHAT OUR GUESTS
ARE SAYING

MAKE A
RESERVATION

ALTASPARQ
AWARE

BEFORE &
AFTER

THE FINISHED
PRODUCT

HOW THE
DISH WAS COOKED

PRAAN
AI

PROBLEM &
SOLUTION

PRAAN AI
DESIGN LANGUAGE

THE FULL
PROTOTYPE

HOW THE
DISH WAS COOKED

KNIGHT
PLAN

BEFORE &
AFTER

THE FULL
PROTOTYPE

KNIGHT PLAN
DESIGN LANGUAGE

HOW THE
DISH WAS COOKED

Web & Mobile:
A Tale of Two Experiences

Four Products.
One Framework.

Nielsen's 10 Heuristics
Across All Platforms

Web vs. Mobile:
The Gap Analysis

4 Key Design
Recommendations

3D PRODUCT
DESIGN

THIS DISH IS
CHEF'S TABLE
ONLY

ALTASPARQ
AWARE

COMPETITIVE
ANALYSIS

HEURISTIC
EVALUATION

SWOT
ANALYSIS

USER
RESEARCH

USER
INTERVIEWS

JOURNEY
MAPPING

IDEATION &
CONCEPTS

HIGH-FIDELITY
DESIGN

USABILITY
TESTING

FINDINGS &
OUTCOMES

THIS RECIPE IS
BREATHWORK ONLY

PRAAN
AI

OPPORTUNITY
IDENTIFICATION

COMPETITIVE
ANALYSIS

USER INTERVIEWS
(21 CONDUCTED)

USER
ARCHETYPES

TARGET
SPECIFICATIONS

PRODUCT
ARCHITECTURE

PROTOTYPE
& TESTING

FINDINGS &
ROADMAP

KNIGHT PLAN
RESEARCH PROCESS

STORY
BOARD

RESEARCH
METHODS

KEY
FINDINGS

IDEATION &
COOL DRILLDOWN

PROTOTYPE
TESTING