Introducing Hush: Receive Notifications Only When They Matter

Shani Mithani
QTMA Insights
Published in
4 min readNov 5, 2018

The Queen’s Technology Media Association provides students access to Canada’s first undergraduate product management program, giving students the opportunity to develop software applications throughout the year to serve today’s market needs.

We’d like to thank everyone for all of the support and feedback received on our previous product showcase. Today we’re going to be showcasing our second product— Hush. Over the last few months, the Hush product team has made tremendous progress towards bringing their idea to life. We’re very excited to be sharing with you what the team has been working on, how they’ve been doing it, and what they have planned for the future. We sat down with Jérémie Bédard, the Product Manager, so we could learn more about the project. Keep scrolling to learn more about Hush— an app that allows you to filter notifications and receive them only when they matter.

So tell us, what is Hush?

Hush is an application which filters push notifications sent to a device to ensure that the user is only alerted with content that is timely and contextually relevant to them. It achieves this by using machine learning to analyze patterns of interaction between the user and their notifications to determine which messages are truly important in the moment.

Why did you want to create Hush?

Everyday, we get bombarded by all kinds of notifications. With dozens of apps on our phones and even more people to connect with, it can be hard to keep track of what is important during busy times. Additionally, being connected 24/7 can distract us from whatever we are trying to focus on in the moment, be it working, studying, relaxing or sleeping. Notifications such as meeting reminders can be important to receive, but we are also flooded with notifications that add no value in the moment and create noise in our lives.

We wanted to create a product which could filter out this noise and allow our users to receive only the notifications that matter to them. Hush is a mobile application that intercepts your notifications from various social networks and only shows these notifications when they are contextually relevant to you.

That’s so cool! Who’s it for?

If you have a mobile device, you can use Hush. Hush can be customized and tailored to your needs. You can customize Hush by asking to only receive urgent messages during meetings, or to turn off all notifications from a specific person. Hush can even understand key phrases from messages to determine whether they should be pushed right away, or if they can wait.

Hush is the perfect product for individuals who lead a busy life, receive a large quantity of messages, value their headspace and want to make sure that it is not disrupted by notifications. Whether you’re a parent, student, or worker, if you want to filter the amount of notifications you receive on a daily basis, you can benefit from Hush.

What’s your tech stack?

Currently, the application is being built using Java with Android Studio and Firebase. Android Studio is the development environment used to build Android applications. Firebase is a Backend-as-a-service (BaaS) platform that can be utilized as a server, database and API. We are using TensorFlow to provide us with an open source machine learning framework.

The first release of the application will be on Google Play.

Awesome! What are the next steps moving forward?

By the end of 2018, we are aiming for a release of a first basic version of the app where the user can set some preferences about the notifications they want to receive. By March 2019, we will have implemented our model that determines the relevance and urgency level of notifications.

The Team Behind Hush

All members of the team are from Queen’s University and a part of the Queen’s Technology Media Association

Thanks for updating us — we’re super excited to see where the team goes with Hush!

Learn more about the Queen’s Technology Media Association at https://qtma.ca/

--

--