LearningPulse
I redesigned the educator flow from document upload to review, helping compress a multi-day manual process into a faster, structured analysis experience while translating unreliable AI outputs into trustworthy UI states.
Role: UX Developer / Product Designer
Timeline: 7 months
Industry: GenAI EdTech
Focus: 0→1 workflow, AI UX, production
6
New NJ District Pilot Signups
From improved website positioning
3
Conference Invitations
NJ + Alaska education conferences
84.6%
Demo Completion Rate
11 of 13 completed full demo flow
The Problem
Educators need AI-generated insights they can actually trust
LearningPulse needed a product experience that could absorb technical variability without making teachers feel lost.
1. Messy inputs
Student documents varied in format, readability, completeness, and quality.
2. Variable AI outputs
LLM- and NLP-based systems could return partial, inconsistent, or low-signal results.
3. Trust gap
Teachers needed to understand what happened, what it meant, & what to do next.
DISCOVERY
What Users Needed Most
5 School Admins / Leaders
3 Teachers
Focus: Qualitative review & workflow pain points
INSIGHT 01
Time pressure was the biggest
pain point
Users did not have enough time to manually
review every artifact in depth.
INSIGHT 02
Value depended on instructional usefulness
Insights had to support real decisions, not just
summarize text.
INSIGHT 03
Trust mattered as much as
speed
Users needed to understand where insights
came from and what to do next.
Designing for the Educator Workflow
The primary personas, school administrators and teachers, needed a system that felt like an extension of their pedagogical expertise, not a black-box replacement. This required balancing technical AI capabilities with familiar, credible UX patterns.
Workflow Transformation
Reframing a Multi-Day Workflow into Minutes
One of the biggest opportunities was workflow compression.
BEFORE
Manual Review Workflow
1
Gather student documents
Collecting physical or scattered digital files from multiple sources.
2
Read each sample individually
Time-consuming manual review of every single submission.
3
Compare patterns across students
Attempting to hold cross-student insights in memory or
scratchpads.
4
Manually synthesize strengths & needs
Drafting summaries based on fragmented notes.
5
Identify next instructional steps
Finally arriving at actionable decisions after heavy cognitive load.
Often takes days
AFTER
LearningPulse Workflow
Upload student work & criteria
Bulk upload digital files or scans into a single unified workspace.
Run AI-assisted analysis
System automatically processes and extracts key themes across the entire cohort.
Review structured summaries & patterns
Clear, scannable dashboards highlighting common strengths and specific needs.
Drill into evidence when needed
Click any insight to see the exact student quotes or work samples backing it up.
Move to instructional decision-making faster
Spend energy on planning interventions rather than organizing data.
Can take minutes
PRODUCT OVERVIEW
LearningPulse is a GenAI-assisted EdTech product that helps educators review qualitative student work, such as writing samples and documents, and turn that material into structured instructional insight.
The platform supports reflection, pattern recognition, and instructional decision-making by helping teachers move from raw student work to organized evidence and next-step insight.
Responsibilities
My Role
I worked at the boundary of product design and implementation in a 0→1
environment. My role was not limited to mockups or handoff — I helped define the
workflow, shape how the system behaved, and ship production improvements.
Workflow Design
Defined upload, configure, analysis, and review flows for educators.
AI UX Translation
Turned LLM, Bedrock, and
spaCy system behavior into
usable UI states.
Product Positioning
Improved website and demo
clarity for prospective pilot
districts.
Implementation
Wrote and shipped production
code in Vue, TypeScript, Python, and SQL.
Context
Constraints That Shaped the Work
Designing a 0→1 AI product required navigating technical realities, unstructured data, and the need to establish trust rapidly with new users.
01
Early-stage product velocity
Most screens were implemented directly in code.
02
Unstructured inputs
Educator-uploaded materials varied in format & completeness.
03
Variable AI outputs
The UI needed stable rendering rules for inconsistent output structures.
04
First-run clarity
New users needed to know exactly what to do next.
05
Edge-case resilience
The product had to preserve trust even when results were incomplete.
Designing for System Logic
Addressing these constraints required mapping UI states directly to backend capabilities. Every design decision was grounded in the reality of what the AI models could reliably produce.
State Mapping
Defined clear visual language for loading, partial success, and error states.
Component Architecture
Built modular Vue components that could handle varying data payloads gracefully.
SYSTEM FLOW
Simplified Architecture
Unstructured Input
PDF, DOCX, TXT
Data Normalization
Python / spaCy
LLM Processing
Variable outputs
Stable UI Render
Vue / TypeScript
Structured Schema
JSON Mapping
Reference Implementation
One hi-fi screen (light + dark) used as reference implementation
Defining the Data → UI Contract
AI output → renderable UI
Creating a reliable contract between backend AI processing and frontend rendering that handles every scenario.
Unstructured inputs
Student work documents uploaded by educators (varied format and quality)
UI-ready insight
structure
sections[] (ordered)
section_title
summary (plain language)
evidence[] with document_id
status (complete / partial /
no_signal)
notes_for_ui (optional)
Consistent UI modules
Predictable cards/blocks in the interface so teachers can scan and drill down
What I owned
Partnered with engineering to translate LLM + spaCy outputs into clear, testable UI states and UI-ready data structures
Defined how insight sections should be structured so the front end can reliably parse and render
Clarified empty/error/partial states so educators always had an understandable next step
Why it matters
Prevents UI ambiguity when outputs change
Makes insights scannable and explainable
Enables consistent iteration without redesigning every screen
Structure in practice
How the data contract translates to UI components
Data Structure
Input
section:
title: "Writing Quality"
summary: "Strong evidence..."
status: "complete"
evidence: [doc_1, doc_3]
UI Component
Output
Writing Quality
Complete
Strong evidence...
Doc 1
Doc 3
Edge state clarity
Every possible state has a clear UI representation
Complete
All evidence found, full summary generated
Partial
Some evidence found, limited summary available
No signal
No relevant evidence detected in documents
Decision enabled
Educators can trust the interface to handle variable AI outputs gracefully, reducing
cognitive load and enabling faster pattern recognition across student work
Designing for resilience: State-Driven UI
Why This Matters:
I treated edge cases as first-class UX requirements because AI workflows rarely follow the "happy path."
The goal was that teachers always understand what happened, what it means, and what to do next.
Design Principle
Implementation Notes
State Management
Vue composables track upload status, analysis progress, and error states
Messaging Strategy
All copy emphasizes user value and next steps, never technical jargon
Component Library
Reusable EmptyState, ProgressIndicator, and
ErrorBoundary components
Testing Coverage
Outcome
By designing for all six states from the start, we reduced "what do I do next?"
confusion and built trust through transparency. Teachers could see exactly where
they were in the process and what to expect.
Information Hierarchy
From "Next Step" → Insight Review
New Analysis (Step-based flow)
Design Process Note
Most screens were designed as mid-fi system blueprints and implemented directly in code to move quickly; one hi-fi screen (light/dark) served as the visual reference.
Reusable UI Patterns
Reusable UI patterns to keep the experience consistent across screens
Key Design Decisions
Step-based Navigation
Breaking the workflow into clear steps reduces cognitive load for first-time users and makes the "next action" obvious at every stage.
Educator-Friendly Language
Avoided technical jargon and data terminology, using terms like "themes," "strengths," and "areas for growth" that resonate with teaching practice.
Progressive Disclosure
Insights are presented as scannable summaries first, with detailed evidence and supporting data available through drill-down interactions.
Consistent UI Patterns
Established reusable components and interaction patterns that work across all screens, ensuring predictability and reducing implementation time.
Market Validation: Website + Demo Narrative
Outcomes from website + demo funnel work
6
New NJ District Pilot Signups
From improved website positioning
3
Conference Invitations
NJ + Alaska education conferences
84.6%
Demo Completion Rate
11 of 13 completed full demo flow
Metrics reflect outcomes from the website + demo funnel work; phrasing should indicate I contributed to / helped drive these results.
What I Did
Owned website + demo experience iteration
Iterated on the website and demo experience in Webflow, refining the product narrative to better communicate value and use cases to prospective pilot districts.
Improved clarity of positioning and flow
Refined messaging and navigation flow so prospects could quickly understand "what LearningPulse does" and how it supports their instructional goals.
Supported pilot momentum and word-of-mouth interest
Made the demo experience easier to follow and more compelling, contributing to pilot signups and conference invitations through clearer value communication.
Before / After Messaging
Narrative direction examples (final copy varied by audience)
Before
Generic, tech-first positioning that didn't communicate clear educational value
After
"Evidence-based insights for student growth"
Outcome-focused messaging that resonates with educators' instructional goals
Impact: Clearer positioning helped prospects understand product value faster, reducing friction in the signup and demo flow.
Outcomes & Reflection
Learnings from data-to-decision products in regulated + GenAI contexts
What Worked
System-level thinking
Stable output structure → predictable UI rendering
Edge-state clarity
Improved trust in AI-assisted workflows
Reusable patterns
Sped up iteration in a startup environment
What I'd Improve Next
Product instrumentation
Lightweight tracking aligned with value signals, not click counting
First-run onboarding
Intentional landing state + quick role-based guidance
Self-serve support
Help center patterns to reduce support bottlenecks
What I Can Demo Live (Sanitized)
Interactive examples and documentation available during interview
Output Schema → Rendering
How structured output enables predictable UI states
No student content shown
Password Reset Flow Spec
Complete requirements + UI state documentation
Production-ready specs
AI Workflow State Matrix
Edge case handling for GenAI workflows
Sanitized examples only
Webflow Page Iteration
Before/after narrative structure improvements
Sanitized content versions
Approach to Portfolio Work
All examples shown are truthful representations of completed work. Metrics and outcomes reflect actual project results without embellishment. Student content and proprietary information have been sanitized or replaced with representative examples.
Verification
References available upon request
Evidence
GitHub + live demos during interview