Soor Malhar

Less friction between signing in and learning music.

An Android app redesign focused on recovering from errors, finding enrolled courses, and understanding the next step in a lesson.

Original project mockup showing redesigned sign-in, home, explore and My Classes screens. — Redesigned core screens. Select any image to inspect it; scroll or pinch in the viewer to zoom.

My role: UX/UI Design Intern
Timeline: May–June 2021
Tools: Figma · Android Studio · Jetpack Compose
AI involvement: None. I did not use AI tools on this project.

My contribution

I explored and refined the interface in Figma and implemented core Android UI screens in Jetpack Compose. Mentor feedback helped me move from adding features to improving everyday usability. The existing app and its learning content were the starting point, not my original product.

Problem

Learning was not the only thing users had to figure out.

The original interface made people interpret errors, hunt for enrolled classes, and distinguish controls from decoration. The redesign focused on those recurring tasks before expanding the feature set.

EnterErrors away from the field
A blocking dialog separated the problem from the input that needed attention.
FindClasses without a clear home
Enrolled courses competed with discovery content on the home screen.
ContinueUnclear interaction cues
A completion indicator resembled loading; non-interactive cards looked tappable.

Existing affinity diagram grouping learner frustrations and business hurdles, including login recall and finding ongoing classes. — Existing synthesis artifact: login recall, finding ongoing classes, and an overwhelming interface. The available record does not document a sample size or research method.

These artifacts informed the design direction; they are not evidence of measured usability improvement.

Decisions

Make the path back to learning easier to recognize.

Four connected decisions follow the core journey: sign in, return to a class, assess a course, and continue a lesson. These are documented design iterations, not results from a controlled user test.

01/04Enter · Recover

Put the error where it can be fixed.

I replaced the all-errors dialog with field-level messages and outlines. Labels and password guidance stay visible so learners do not have to remember instructions while typing.

A blocking dialog interrupts sign-in and lists errors separately from the fields.

Inline messages identify the specific fields that need attention.

Iteration: I removed the extra confirm-password field from the signup exploration to avoid repeated entry and retained password visibility control. Social sign-in remained a feature suggestion, not a delivered integration.

02/04Find · Return

Give enrolled classes their own destination.

My Classes replaced notifications in the bottom navigation; notifications moved to the top bar. Recent lessons sit above the enrolled-course grid, making resuming and browsing distinct tasks.

Revised navigation gives enrolled classes a direct destination.

A horizontal Recent row leaves vertical space for the My Courses grid.

One destination, two intentions

Resume: recent lessons are grouped first.

Browse: enrolled courses remain visible below.

The earlier list layout was revised because its course images were not working well in that format.

Trade-off: move a less central destination out of the bottom bar without removing access to it. Home can focus on discovery instead of duplicating enrolled courses.

03/04Assess · Choose

Let the course content carry the hierarchy.

I grouped course facts, learning goals, and the description into a clearer reading order. Removing card-like styling from static information reduced false affordances; reviews retained their interactive treatment.

The original overview divides course information across visually heavy cards.

The redesign groups course contents, learning goals and Description above the start action.

Refinement: I tried an orange start action in an earlier visual iteration, then returned to the brand color. Ratings, student counts, and other screenshot values are example course content—not redesign outcomes.

04/04Continue · Recover

Make the next action unmistakable.

In lessons, consistent Previous and Next controls replaced inconsistent navigation, and I removed a completion indicator that looked like loading. The same principle shaped the offline state: one clear recovery action.

Overlapping text and a prominent graphic compete with Try Again.

A subdued illustration leaves a clearly outlined Try Again button as the action.

Across the lesson flow: I also moved comment composition away from a blocking dialog toward an input attached to the keyboard. These are documented interaction decisions, not validated completion-rate gains.

Impact

Core flows redesigned. Native UI built.

The project delivered design iterations and Android UI implementation—not a measured improvement in engagement, learning, or task success.

Design delivery: The existing project record lists approximately 20 explored or redesigned screens. The source document preserves original screens, intermediate versions, and final design decisions.
Android UI: The existing record lists 8 core screens implemented in Jetpack Compose. This describes UI work, not a complete backend integration or a verified production release.
Testing & outcomes: No documented participant count, before/after usability measurement, or verified engagement lift is available. I do not attribute product-impact metrics to this redesign.

Explorations beyond the core journey

The project also explored a wishlist, saved posts, course FAQs, progress reporting, and completion feedback. Social login, voice search, and XP-based ideas were proposals rather than verified shipped functionality.

A course-level FAQ destination proposed alongside the learning flow.

A completion concept with a clearer next-session action; XP was a proposal.

What I took forward

More features did not automatically make a better learning experience. The useful shift was to clarify errors, navigation, and action cues first, then carry that intent from Figma into native Android UI.

Explore the original design record