MemQ: Smart Quiz & Memory APP
A streamlined mobile learning experience designed for lifelong learners to capture knowledge and master subjects through custom flashcards and quizzes.
The Design Problem
Most learners only have fragmented 5–10 minute windows to study, so the interaction needs to feel immediate and frictionless. My research revealed that card creation is the make-or-break moment where a habit is either formed or abandoned—especially when users are commuting, typing on a mobile keyboard, or trying to capture a thought before it fades.
Key Pain Points
Irrelevant Content
Pre-made decks force users to waste time on concepts they already know, rather than their specific knowledge gaps.
Creation Friction
The complexity of manual entry turns capturing a quick thought into a chore, often leading to abandonment.
Static Scheduling
Without smart prioritization, critical and urgent concepts get lost in a sea of easy, linear tasks.
What I Designed
I designed MemQ to support every learner's pace: the crammers, the casual reviewers, and the creators. The app keeps users focused on retention while giving them complete control over how they capture fleeting ideas and structure their personal knowledge base.
I designed an AI-Capture System that eliminates data entry chores
My research showed that the biggest barrier to habit formation was the time spent manually creating cards. Users often spent more energy formatting text than actually learning. I solved this by building a multi-modal AI engine: users can simply upload a lecture PDF, type a topic, or chat with the assistant, and MemQ automatically extracts key concepts.
Usability tests revealed that context matters. A vocabulary word needs a definition and example sentence, while a complex concept needs a "Why" or "How" question. So, I engineered the backend to detect the content type (Knowledge Point vs. Vocabulary) and generate the most effective question format for that specific item.
Key creation features:
- Multi-modal Input to accommodate any source material (Manual, PDF Upload, Topic Generation).
- Context-aware Extraction to automatically turn chat conversations into saved flashcards.
- Adaptive Question Generator that formats quizzes differently for vocabulary (definitions) vs. concepts (conceptual understanding).
Design Responses
Creation friction
Multi-modal AI input (PDF, Chat) + auto-formatting
Irrelevant content
Context-aware generation tailored to specific knowledge gaps
Static scheduling
Dynamic Spaced Repetition algorithm + Smart priority queue
Context-Aware Engine that structures knowledge
The AI engine intelligently detects the content type—distinguishing between vocabulary and complex concepts—to generate the most effective question formats. It transforms passive reading materials into interactive quizzes without the user needing to write a single prompt.
Used for: Concept extraction, smart formatting, instant definitions
Impact
Usability testing revealed that the new "One-Tap Capture" flow significantly reduced the barrier to entry. Where users previously abandoned manual entry after 30 seconds, the AI-assisted design enabled them to complete the loop in under 5 seconds.
Testers praised the "invisible interface" strategy—where complex tasks like tagging and syncing happened in the background—allowing them to focus entirely on learning. This shift in UX directly contributed to higher retention rates and validated the transition to a premium subscription model.
What improved between rounds
Critical issues
Round 1 surfaced high abandonment rates during manual entry. The AI-assisted update resolved this friction point completely and had 0 critical drop-offs in the final validation.
Flow & confidence
Participants described the AI capture flow as 'magic, efficient, and intuitive' and said they felt 100% confident that the app correctly understood their study materials.
Engagement shift
Instead of complaining about typing fatigue, round two feedback shifted to requests for more file formats, showing users had moved past the friction of starting to active daily use.
Data-Informed Design Decisions
Preferred method of card creation
Aligning with this, I prioritized AI generation as the primary action, relegating manual entry to a secondary option to minimize friction.
Importance of personalized study content
This drove the decision to build a context-aware engine instead of a marketplace, ensuring quizzes are generated directly from the user's own materials.
"This is the most seamless study experience I've seen. I would feel 100% confident ditching my old messy notes for this."
— Round 2 usability testing participant
Future Improvements
Smarter multi-modal capture
Use desktop browser extensions for PDFs and fast capture. Future integrations with major LLMs, Notion, and camera-based capture will let learners save key concepts from chats, notes, and real-world materials in a single step.
Richer multi-modal recall
Add richer multimedia support—starting with images captured from the camera—to reinforce memory at both the term and question level, tying abstract concepts to concrete visual cues.
Community-sourced patterns
Explore class-style cohorts where learners can keep each other accountable while selectively sharing materials and deck patterns that others can remix.
Dig Deeper
Personas
Meet Alex and Taylor
Used to capture different learning motivations: Urgent Exam Cramming , and Casual Learning.
Affinity Map
Synthesizing research insights
How I clustered interview data into three core friction themes: Manual Entry Fatigue, Content Relevance, and Scheduling Guilt.
User Flows
Complete user journey flows
Visualizing the complete user journey across capture, learning, and study workflows.