Share
Explore

icon picker
Explaining UI

​
Gavriel Sharabi
Presenting 4 projects to prove how understanding product goals and my UI-specific decisions impacted people and business needs.

Ask.uni

A student led advising app


1. Project Overview 🌈

For students, by students. Designed an advising and Q&A app for students looking for relevant college advice from university peers and exciting academic relationships.
I developed the IA, content strategy, and MVP prototype for the app.
⁠
SuccessMetrics.png
⁠
⁠

2. The Story 📓

While initially sketching an idea of mine, I was contacted by I was contracted by a group of student entrepreneurs at UCLA to create a low-budget MVP prototype for them to base and validate their designs for their new advising platform called Vize—the following app is the result.
A solution tailored to the 2021 fall freshman class.

3. Problem & Solution 🏃‍♀️

Tailoring the app to the Gen Z audience, I had to come up with a content strategy that would most appeal to users of the high engagement-fast returns economy. Also, needing to design for a post-pandemic world, I decided to create this app in a manner that allows for real connections, despite the ongoing social-distancing IRL. By taking cues from popular community engagement apps like reddit and discord, I have created an app where new students can connect with other students for advice, mentorship, and relationships in private or in a group according to their academic goals and interests.
⁠
SuccessMetrics-1.png
⁠
⁠

4. Results ✈️

Testing the product with users and discovering sub-par competition in the market has led me to positive results:
More than 800 freshman students answered that they would love to access this platform that would help them navigate their university life.
Advisors from my university (FIU) commented that this app would reduce their overhead time, allowing them to better service students on particular university goals.
Over 300 experienced students highlighted that they would like to mentor newer students and get a habit of building industry based networks.
On to next steps: the team at Vize is currently in beta development of their own app based on Ask.uni’s designs, using it as a source of truth for their proof-of-concept. On the flip side, I will be doing some no-code development to launch this project around Spring 2022. Who's onboard with me?
View Ask.uni
⁠
⁠

Albert

A book e-commerce site...written & published by a dog.


1. Project Overview 🌈

Designed a website to help promote and sell a new book by an author friend of mine—detailing human life by a dog’s POV. Within 2 sessions we created together a brand & marketing strategy to ensure maximum exposure, and decided on a visual design direction for her site. My main on this project roles were strategy, visual, and web design.
⁠
סטיקר חלול.png
⁠
⁠

2. The Story 📓

The book was published during the start of the second wave of Covid-19 lockdown restrictions faced in Israel, where the only time people could leave their house was for essential work, sports activities, or grocery shopping. And after printing 500 copies for the first sales batch, this was a major hurdle for the inexpensive and organic exposure we had hoped to reach.

3. Problems & Solution 🏃‍♀️

We strategized on capturing people’s attention with the minimalist graphics of her book cover—a title and a dog illustration. However, we quickly understood that outside of her regular fanbase, even avid Israeli readers were mostly unfamiliar with her book and that just publishing a site would not reach new readers. Also we knew that running sponsored ads to promote her website would prove expensive and would lose the mystery of the brand direction we thought would convert the attention of interested viewers to proud owners of Albert as viewers would only see a promoted link or just keep scrolling through their social feeds.
So we took a radical approach by creating a website that sticks to the minimalist context with only a bit of copy, examples of social proof, and book pages to further intrigue leads and made a QR sticker that leads to the book’s website, and plastering 500 copies of it all over TLV’s Covid-friendly hotspots. Nearby essential workplaces, sports areas, and grocery stores.
⁠
Untitled design.jpg
⁠
⁠

4. Results ✈️

Within the first three weeks, we managed to sell +320 books, with all 500 of the first book batch being sold before the end of the month. Working on this project gave me a chance to shine by bringing breadth of experience in project, content, and operations strategy to leverage my digital designs, and successfully ship a physical product. Pick up some Hebrew, and learn what a dog has to say about all of this!
View Albert by Noalle.
⁠
⁠

Hyg

An affordable co-living finder app


1. Project Overview 🌈

Mapped a proposed solution to finding affordable housing during the Covid-19 crisis. In 2 weeks of design sprints I envisioned a market fit for the product and diving into a mix of both qualitative and quantitative research methods to design an app that allows users to find Co-living spaces to rent, book, and share.
⁠
Desktop - 4.png
⁠
⁠

2. The Story 📓

There's so much to improve in the home finder industry. Even during the height of this pandemic, people who looked to move had difficulties finding affordable housing.
Tech, and global catastrophes are pushing alternatives to traditional housing. Among those, Co-living is becoming increasingly popular.

3. Problems & Solution 🏃‍♀️

What people struggled with:
Financials: People not having enough money to afford standard renting options due to Covid-19.
Discovery: Finding alternative renting spaces through conventional apps and websites like Facebook and Google, is difficult. Co-living spaces are especially hard to find as most are advertised in a decentralized manner. (WhatsApp groups, poor-SEO websites)
Social Impact: Humans impacted the most are young professionals, as well as service, restaurant, & hospitality industry workers.
Solution: An Android app for users to easily see information relevant to their needs and book a living space.
⁠
Hyg.png
⁠
⁠

4. Results ✈️

Working on this project was challenging as the concept for an app like this is rare, as well as learning how to design an Idea from scratch. This project made me realize the value of creating relevant designs for user engagement. In the future I hope to take the skills I developed for this project to craft memorable, relevant, and consistent user experiences for people affected during financial crises.
View Hyg.
⁠
⁠

My Daily Nosh

Recipe and grocery shopping list personalize w/ image recognition & ML


1. Project Overview 🌈

Leading the design work for a client wishing to create a grocery shopping app that uses
Machine Learning
⁠
to order groceries and suggest recipes based on the user’s preferences, budget, and order history.
⁠
screencapture-gavrisharabi-design-work-mydailynosh-2020-12-21-17_23_09.png
⁠
⁠

2. The Story 📓

My time constraint to learn relevant machine learning practices and finish the visual designs was 2 days:
The 1st for sketching, wireframing, and designing.
The 2nd for testing and refining.

3. Problem & Solution 🏃‍♀️

Not knowing anything about ML, I started researching ML patterns and best uses in mobile UIs. Material Design’s guide for image recognition served as the base for my image capturing screens.
⁠
My Daily Nosh.png
⁠
⁠

4. Results ✈️

This design challenge started with a 2-day deadline for completion. I feel that this challenge made me more agile in my visual design process as I pushed through to flow and sketch for the first day, and tested to iterate for a finished design on the next.
It surprised me to see the satisfaction and excitement that personalized content gives to users. Relevant design through machine learning allows users to feel more connected to certain apps and brands, making designs not only extremely helpful to customers but also very valuable to brand shareholders as increased satisfaction, increases engagement and revenues.
View My Daily Nosh.
⁠
⁠

Want to print your doc?
This is not the way.
Try clicking the ⋯ next to your doc name or using a keyboard shortcut (
CtrlP
) instead.