01
Heading
Lorem ipsum dolor sit amet, consectetur adipiscing elit. Suspendisse varius enim in eros elementum tristique. Duis cursus, mi quis viverra ornare, eros dolor interdum nulla, ut commodo diam libero vitae erat. Aenean faucibus nibh et justo cursus id rutrum lorem imperdiet. Nunc ut sem vitae risus tristique posuere.
Lorem ipsum dolor sit amet, consectetur adipiscing elit.
01
Lorem ipsum dolor sit amet, consectetur adipiscing elit. Suspendisse varius enim in eros elementum tristique. Duis cursus, mi quis viverra ornare, eros dolor interdum nulla, ut commodo diam libero vitae erat. Aenean faucibus nibh et justo cursus id rutrum lorem imperdiet. Nunc ut sem vitae risus tristique posuere.
Heading
Lorem ipsum dolor sit amet, consectetur adipiscing elit.
Heading
Lorem ipsum dolor sit amet, consectetur adipiscing elit. Suspendisse varius enim in eros elementum tristique. Duis cursus, mi quis viverra ornare, eros dolor interdum nulla, ut commodo diam libero vitae erat. Aenean faucibus nibh et justo cursus id rutrum lorem imperdiet. Nunc ut sem vitae risus tristique posuere.
Kuri
Introduction
Kuri is a personal assistant powered by AI. It uses large language model (LLM) technology, allowing it to understand and respond to your requests in a more intuitive and natural way. Kuri is designed for collaborating on projects, answering queries of various levels of complexity, and generating content for most productivity related tasks.
Kuri learns its users' preferences and tailors its responses to their unique style--it's the perfect tool for anyone looking to streamline their daily routine and maximize their productivity. This idea came from our team's observations about other people's reactions to newly developed digital tools powered by artificial intelligence.
Kuri
Challenges
- Design an interface that leverages users' schema and prior knowledge.
- Create ways for users to customize the generative experience.
- Enable users to save and retrieve generated content with minimum effort.
Kuri
Method
In this project we utilized the Goal-Directed Design (GDD) process, a product design methodology developed by Alan Cooper, which focuses on providing design solutions that meet users’ goals and needs while still addressing business/organizational requirements.
In the following sections you will find a walkthrough of each phase of the Goal-Directed Design process leading to the final prototype. In each phase (Research, Modeling, Requirements Definition, and Refinement), respective documentation is provided in order to demonstrate the insights, challenges, and design solutions developed by our design team. This process was performed as a class project being conducted as part of the requirements for TCIAD4700 (Capstone Project and Portfolio Showcase) at Kennesaw State University, and supervised by Dr. Leslie Hankey.
Kuri
Research
The Research phase of the Goal-Directed Design is focuses on gathering qualitative data and relevant information on the subject, domain, and potential users as it relates to the product. This phase had four major steps: Kickoff Meeting, Literature Review, Competitive Audit, and User Observation/Interviews.
Besides assisting the design team during User Observation and Interviews, I was tasked with performing a Competitive Audit of products that used generative AI technology.
Usually, in Kickoff Meetings, stakeholders introduce the project idea to design and development teams. Designers ask questions about users, competitors, and potential challenges. For our class project, the design team brainstormed and collaborated on answering fundamental questions based on assumptions to create a problem statement and assumptions statements to guide our process.
Kickoff Meeting
Kuri design team's Kickoff Meeting.
Literature Review
The Literature Review give designers the opportunity to explore relevant literature related to the product to be developed and its domain. This step is important because the resulting material is used as the basis for better informed stakeholder and subject matter expert (SMEs) interviews.
My teammates Andrea and Abby performed our Literature Review and introduced us to the following insights:
- Public understanding of AI is broad but shallow; people often encounter AI without realizing it or understanding its applications.
- Young people are most optimistic about AI and are more likely to embrace automation in the workplace.
- People expect virtual assistants and chatbots to treat its users with sympathy and care (humanized technology).
- Misinformation is a huge concern for many AI users.
Competitive Audit
In the Competitive Audit phase, the design team have the chance to review the systems, interactions, and interfaces of competitive products already established in the same domain. This step gives designers an insight into the strengths and limitations of certain features, as well as trends and design patterns of comparing products on the market.
For project Kuri, I joined Jhordan John and Kinsey Still in exploring the competing products in the domain of generative AI such as ChatGPT and Midjourney, and common search engines like Google and Bing. Our findings helped us understand the strengths, weaknesses, and above all, the opportunities that our team could explore in bringing Kuri to life.
Project Kuri's Competitive Audit documentation.
User Research Interviews
In Goal-Directed Design, we use the technique of ethnographic interviews to gather valuable qualitative data to help us better understand users and their goals. This technique combines immersive observation and directed interview, and the focus lies on accessing interviewees’ contexts in order to better understand their attitudes, beliefs, and values, as it relates to the product’s domain.
We interviewed and observed six user research participants from varying ages, and asked questions that ranged from their perception of AI, to their preferences when searching for and learning new information. As a team, we learned that participants were already integrating AI technologies into their workflow, and their main concerns revolved around the validity of the information generated, issues with biases, and the technicalities involved in storing and retrieving content.
Project Kuri virtual User Research Interview conducted through MS Teams.
Kuri
Modeling
The resulting synthesis of the information and qualitative data gathered during research takes shape in the form of Personas in the Modeling phase of GDD. Personas represent the synthesized behavior patterns, motivations, attitudes, opinions, and mental models observed during user research.
Through relevant abstraction, Personas serve as a tangible and guiding user model for which the product will be designed. A successful product accommodates a variety of users by being designed for specific types of individuals with specified wants and needs.
Post User Research Interview Affinity Mapping.
Visual Continuum Matrix
Once Research concluded, we engaged in critical discussion and collaboration to synthesize our data and construct our models. We identified the behavioral variables observed during user interviews and analyzed during post-interview Affinity Mapping sessions. Then we checked for redundancy and traced back each variable to a validating signal in our data.
Lastly, we assigned each listed variable one of two classes: single continuum variable (more/less), or multiple choice variable (this/that). We then mapped interview subjects to listed behaviors by assigning each one a number (1-5) and plotting them in our visual continuum matrix of behavior variables.
Visual Continuum Matrix of behavioral variables.
Personas
We identified significant behavior patterns by observing clusters of represented user subjects. In other words, numbers grouped together indicated a strong behavior pattern. We synthesized the characteristics of the key behavior continuum for each of the two identified Personas and then defined their goals. We also gave them names and a picture so as to humanize their represented personalities.
Primary Persona
Name: Daniel Price
Age: 30
General info
- Works as a product manager.
- Lives in New York City.
- Interested in new technologies.
- Likes to learn new things.
End Goals
- Wants to be able to execute specific tasks as efficiently as possible.
- Wants to be able to personalize their experience based on tasks they are completing.
- Wants to find a way to streamline how they store and manage information.
Secondary Persona
Name: Kim Stewart
Age: 23
General info:
- She is a graduate student.
- Lives in Miami.
- Enjoys relaxation with her creative outlets.
- Likes to support small local businesses.
End Goals:
- Wants to customize their experience without violating their privacy.
- Wants to explore the new technology in a familiar environment.
- Wants to have a creative tool that allows them to explore personal interests.
Kuri
Requirements
The Requirements phase of GDD is perhaps one of the most crucial steps in the design process. In this phase, designers translate their research and empirical observations into design solutions that satisfy user needs, while also maneuvering technical constraints and addressing business goals.
Creating Context Scenarios involved the use of narrative as a design tool to contextually describe the user’s interaction with the system of our product. These were done in the form of a short narrative story of a day in the life of the Persona, while describing, at high level, the actions and expectations they have with the product.
Kuri
Frameworks
In the Design Framework phase of GDD, we set up the structure of our product’s system interface through the organization of interactive behaviors, visual language, and functionality. This process began with the Interaction Framework and the creation of low fidelity wireframes, followed by the Visual Design Framework and the establishment of a visual style and identity.
Project Kuri's initial wireframe layout.
Kuri
Refinement
The Refinement phase marks the transition between low-fidelity wireframes to a high-fidelity and fully functional prototype. At this stage, the refinement of form and behavior translates into full-resolution screens representing the user interface. In other words, it is during the Refinement phase of GDD that the product comes to life both visually and interactively.
Kuri
Takeaways
The creation of Kuri had many challenges, such as the research that was necessary to understand a new technology and interpreting the behaviors of user research participants interacting with it. Despite all the novelty, users of new technologies bring with them certain pre-conceived notions, assumptions, and expectations as to how a new system should behave. Leveraging these existing mental models offers a lot of potential for getting users up to speed with a new product.
If given the opportunity to revisit project Kuri, I would have chosen to usability test the final prototype with users by introducing card-sorting as a way to better understand their expectations regarding the information architecture of menus and the labeling in our product's microcopy.