Front end · UX · Boston University

Playlist Maker

A web app that turns your trip into a Spotify playlist: sized to how long you’ll be traveling, built from your library, and matched to the weather.

view on GitHub
Playlist Maker transport screen asking How are you getting there? with Walking, Biking and Driving buttons

Role

Front end and design, on a team project

Course

CS411, my final computer science course at Boston University

I learned

HTML, CSS and React, during the semester

Stack

React, Flask, Redis; Spotify, OpenWeatherMap and Google Maps APIs

The brief

A semester-long project with four technical requirements. My idea: a playlist that lasts exactly as long as your trip.

Good weather adds more upbeat, happy songs. Rain, snow, clouds or cold bring in mellower ones.

Redis databaseRequired data store
Two APIsOpenWeatherMap for the weather, Spotify Web API for the playlist
Third-party loginSpotify sign-in with OAuth 2.0
Decoupled architecturePython Flask back end, React front end
Extra: Google MapsGeocoding to find each location, Distance Matrix to calculate the trip time

01 · First draft

Start plain, then shape the flow

I started with a basic page: “Welcome [name]”, then fields for destination, current location, mode of transport, desired speed, and music or podcast. A template page helped me understand layout before I designed anything.

First draft: a single form with text fields and dropdowns for transport, speed and music or podcast
The first single-page form
Close-up of the speed dropdown with Slow, Normal and Fast options
Dropdowns hid the choices behind extra clicks

02 · Sketching in Figma

One question per screen

In Figma I sketched what I had in mind: a destination search with a map, then one screen each for transport and speed, ending on the playlist.

Dropdowns for transport and speed became big buttons on their own screens. That cut clicks and made each choice obvious at a glance.

Figma file with four screens: destination with a map, speed buttons, transport buttons and the finished playlist
Figma sketches: destination, speed, transport and playlist

03 · Decisions

What we changed, and what we cut

Origin to destination. I reordered the inputs to read logically, and we used Google Maps to work out the distance and trip time.

No podcasts. Episodes run from 5 minutes to 2 hours, while songs are 3–4 minutes. As BU students we tested trips on campus, and the longest took under 30 minutes, so songs fit far better. As a team we dropped podcasts and focused on song playlists.

3–4 mina typical song
5 min – 2 hra podcast episode
< 30 minthe longest trip across BU’s campus

04 · Visual design

Spotify colors, checked for access

Since the music comes from Spotify, I built the palette from its brand colors: white, green and dark grey. I checked the swatches in Adobe’s accessibility tool, which found no conflicts and confirmed they are color-blind safe.

For type I wanted something close to Spotify’s Gotham: clean, modern, fun and approachable. I chose Poppins, which is also the font on this site.

Adobe Color accessibility check showing five swatches with no conflicts found and a color-blind safe confirmation
The palette passed Adobe’s color-blind check
#DCEAE7
#2F5451
#1DB954
#96CC95
#F2D1EF

Poppins

Aa Bb Cc 0123 · clean, modern, approachable

05 · The final product

From login to playlist

Users log in with Spotify, enter where they are and where they’re going, then pick how they’re getting there and how fast. The app builds a playlist that matches the trip length, from their own library and tuned to the weather: upbeat when it’s sunny and warm, mellow when it’s rainy, snowy or cold.

Playlist Generator landing page with a Get Started button
Start: log in with Spotify
Destination screen with current location and destination fields
Where are you going?
Transport screen with Walking, Biking and Driving buttons
How are you getting there?
Playlist Generated screen showing the playlist's album art, a My Playlist button and a Start Over button
Your playlist is ready, with one tap to open it in Spotify
The generated playlist open in Spotify: 10 songs, 36 minutes 38 seconds
Saved to the user's Spotify: 10 songs, 36 min 38 sec

Under the hood

The full code for Playlist Maker is on GitHub.

GitHub repo more of my work →