Custom Preset Copy 11.png


Zach Bachiri is an interaction designer on a mission to create social change through design and technology.



CMU M.Des. Interaction Studio 2, 12 weeks

Team Members

Devika Khowala, Hajira Qazi, Shengzhi Wu


Sketch, Principle, After Effects



Although I am well-traveled and I expected to know more about the US, I just get surprised every time a new thing is thrown at me, and I go quiet because I don’t know how to react.
— Carnegie Mellon International Student

Each year, there 1.18 million international students in the United States. Many of them go through a process of acclimation that can be, at times, stressful and overwhelming.

During our M.Des. Studio 2, which attempted to explore the connection between artificial intelligence and learning, we attempted to design a solution that would create a more inclusive environment for international students on campus. We framed out design challenge as follows:

How might we use artificial intelligence to aid cultural learning for international students in the United States?

We followed a process with 3 phases of research: exploratory, generative, and evaluative. As a result, we were able to experiment with a variety of research methods, which led to some valuable insights and concept directions.


Exploratory Research

We began by mapping the territory of the problem space. We looked at the types of immigrants experiences issues of cultural learn, what types of things they need to learn, how they go about learning those things, and then possible design framings.

Territory Map 5.png

We then conducted exploratory research with international students on campus to better understand what pain points they encountered and how they overcame them when first arriving to the US. We used a variety of research methods to get at these questions. Methods like interviews, surveys, journey mapping, storyboarding, and letter writing were leveraged.


exploratory research workshop

research preparation

participant journey maps

Through this research we discovered some initial insights that guided us moving forward:

1. People largely rely on social interaction as a way of adapting to a new culture. Community and a social network are key to this process.

2. People can’t prepare for every situation. Sometimes they need help in the moment of an interaction they could not have prepared for.

3. Because the topics of learning are so varied, resources for immigrants are fragmented and hard to find.


Generative Research

After our exploratory research, we began generative research with the goal of sparking ideas and concepts around how international students can overcome the various hurdles we heard about earlier.


We held a two part participatory design sessions in which we identified the most common problematic scenarios international students come up against and then generated solutions to those scenarios. With these solutions we were not looking for 1-to-1 solutions, but rather looking for themes that emerged which would inform our final design.


scenario building workshop

solution building workshop

participant output example


We generated the following insights through participatory design workshop:

1. A bi-directional cultural learning process makes the acclimation process for immigrants easier and more effective than unidirectional approach.

2. International students try to avoid unfamiliar situations they anticipate to be difficult, and will go out of their way to do so.

3. Having a supportive community is important to the mental and emotional wellbeing of international students.


We then developed 3 concept directions to speed date with international students. We received positive feedback about the social network with facilitated interaction and decided to move forward with this concept. There were a few reasons for this. First, and most importantly, a network would allow for us to organically create bi-directional learning between international and American students. Second, this platform created an incentive for both international and American students to join. And finally, by creating an intervention before students arrive on campus, we are able to create a more welcoming environment before students even arrive in the US.


Evaluative Research

With our concept direction set, we began formalizing the user journey and creating wireframes. We developed various iterations of the key features and began gathering feedback from users.




AI assistant



We focused much of our testing on the chat intervention and matching experiences, as those were integral to our concept and where artificial intelligence is leveraged most.


chat intervention iterations for testing

matching iterations for testing


We received vital feedback from our testing that we then integrated into our final designs.


1. People didn't like when the bot made assumptions about them or put words into their mouth.

2. People liked having multiple options provided to them, as they were unsure about the accuracy of the bot intervention.

3. People preferred having more information about their matches, but were actually okay with having anonymized matches. Some saw the value in not the anonymity of matching.


Conversation Analysis

To better understand how the facilitation bot would function during a conversation, we also conducted conversation analysis on conversations that we setup between incoming CMU Design students. We printed out these conversations and noted points of breakdown or missed opportunity.

conversation analysis

conversation analysis


We discovered that breakdowns didn’t happen as often as expected, but there were lots of opportunities cultural learning that were not capitalized on. That shifted our thinking about the bot from being about fixing breakdowns to proactively creating opportunities for cultural learning as a way of increasing engagement between students.


Naya Platform

The result of all this research and iteration is the Naya platform for incoming students. Naya is a social networking platform for incoming students that leverages AI to facilitate bidirectional cultural learning through person to person conversation.


When students enroll at a university, the school will invite each incoming class to join this platform prior to arriving on campus. By creating an intervention that allows students to begin cultural learning before stepping foot on campus, we hope to lessen the effects of cultural shock by providing international students with access to key knowledge and creating a support network prior to arrival.



The Naya onboarding experience creates expectancy of a positive social experience and cultural learning with their classmates. It is also an opportunity for students to indicate interests that will later be used to match with other students and make recommendations.


Main Navigation

Direct Messages

Groups Messages

Group Discovery

Settings & Profile


Naya Bot

The goal of Naya bot is to provide a unified resource for students to access university information. In our research, we heard about the lack of centralized resources and information for incoming students. This attempts to address that by unifying various university resources into a single source.



Matching attempts to initiate relationships between international and American students on the Naya platform and remove implicit bias in these connections. Through a nameless and faceless matching system, students connect based on their interests, not stereotypes.


Chat With Bot Facilitation

Chat encourages international and American students to talk about topics that foster bidirectional cultural learning. The suggestion feature recognizes topics of conversation and surfaces topics with prewritten messages that will hopefully spark cultural learning through conversation.


If the bot recognizes a potential topic of conversation but is not able to craft a prewritten message, it surfaces a prompt to the user for them to write their own message.


Bookmarking allows users to save and review information to refer back to later. These saved messages are automatically sorted by topic. By doing this, students will be able to create an index of information they can go to when they encounter issues on campus or want to reference back to something that arose in a conversation.


A.I. Knowledge Graph

To make these suggestions, we will need a knowledge graph that draws connections between topics. As students act upon our suggestions, connections in the graph become stronger and suggestions become more confident.