This is just a display page and the full case study is coming soon.
Neuron is e-GMAT's official-question practice platform. I designed it end-to-end - design system, components, and every screen - so students stop juggling platforms to solve official GMAT questions.
Students preparing for the GMAT need to practice on official questions. But no platform in the market combined those questions with proper analytics or quality solutions. Students were left juggling GMAT Club, the Official Guide, and other scattered sources — getting the questions, but no visibility into where they were actually stuck.
The platform-juggling wasn't just inconvenient. It broke the practice → review → re-practice loop students needed in the final weeks of prep. Switching tools meant losing context, mistakes weren't tracked across sources, and methodology was always one tab away from the question that needed it.
Students don't study in a line — they solve, review, bookmark, return, retry under the timer. Neuron treats questions, sessions, timed and untimed modes, and bookmarks as one continuous loop, not separate features.
A doubt gets resolved where it arises. The e-GMAT methodology lives inside the solution view, and the forums are one click from the question — "I don't understand why" is answered without leaving the screen.
Every platform had the questions. None had the loop — practice, review, doubt-clearing, retry — without switching tools. Neuron is the workspace where that loop lives.
I designed the product end-to-end. I built the design system, the component library, and every screen in Figma, then handed the files to engineering for implementation. I did not write code; every visual and interaction decision in the product is mine.
Neuron is built on two principles from the founding brief: no platform-juggling, and practice OG questions the right way. The design translates those into five concrete choices a student feels every time they open the product.
The custom-quizzing flow walks a student through section -> topic -> subtype -> difficulty -> source -> question pool, then surfaces the matching question set. Each step narrows scope, so a student lands on the right thirty questions instead of staring at three thousand.
The first screen narrows a broad question universe into a guided setup flow.
Neuron let me drill official GMAT questions on my exact weak areas, reducing my solve time while improving accuracy.
At the question level, "Start Timer" and "Show Solution" are equal-weight actions. Students choose whether they are attempting under exam conditions or reviewing for understanding - without burying either path in a menu.
The proprietary frameworks - passage analysis, pre-thinking, answer-choice elimination - live inside the solution view of every question, not in a separate course. Methodology gets taught at the moment of error, where retention is highest.
Method, timing, and review all live inside the same solution surface.
This targeted approach drove my CR hard accuracy from 60% to 82%.
History is not a log. The attempts view filters by correct, incorrect, and bookmarked, links straight back to full solutions, and feeds back into custom quizzing - closing the loop between practice, review, and re-practice. Final-week prep was a core scenario in the design brief.
The dashboard starts as a sortable attempt history with clear review entry points.
Identify weakness → study module → practice official questions on Neuron, retest — created measurable improvement between each mock attempt.
Tokens, primitives, components, layout templates - built in Figma so the team could ship the current platform and plan its v2 expansion (Data Insights questions, AI-generated similar questions, deeper analytics) without rebuilding the foundation.
After launch, public feedback on GMAT Club and Reddit started landing on the same things the product was designed to do — measurable score lifts traced back to specific behaviors, targeted weak-area drilling, and a practice loop tight enough to use between mock attempts.
Quotes below are verbatim from publicly posted reviews and debriefs on GMAT Club and Reddit. Emphasis is mine, to mark the part most relevant to the product design. Each quote links to its original source — these are public observations, not solicited testimonials.
When I had specific gaps in hard questions, Neuron was perfect for targeted practice. I used it to fill knowledge gaps without time pressure.
Neuron helped me identify consistent patterns in my mistakes and fix the behavioural errors that I had identified.
I started using Neuron for 15-20 minute warm-up sessions before cementing, gradually building my attention span.
When I had specific gaps in hard questions, Neuron was perfect for targeted practice. I used it to fill knowledge gaps without time pressure.
Neuron helped me identify consistent patterns in my mistakes and fix the behavioural errors that I had identified.
I started using Neuron for 15-20 minute warm-up sessions before cementing, gradually building my attention span.
