Topper Advisor
Conversational AI for EdTech
An AI chatbot that streamlines elective selection and tailors degree pathways for university students, easing the advising workload.
Client
St. Edward's University
Timeline
10 Weeks
Season
Spring 2024
Focus Areas
ed-tech
prototypes
wireframes
user-research
ai
personalized-design
Tools Used
Figma
Miro
HTML & CSS
Python
Flask
project overview
project summary
Topper Advisor was developed to simplify the elective selection process for students at St. Edward's University. Utilizing a conversational AI, the chatbot aids students in finding courses that match their schedules and academic goals by providing detailed course information. The project aimed to improve user satisfaction and reduce the workload of academic advisors.
role summary
As UX Designer, UX Researcher, and Developer, I led the UX design, contributed to UX research and prototyping, and developed the chatbot using HTML, CSS, Python, JavaScript, Flask, and OpenAI’s Assistant API.
project summary
Topper Advisor was developed to simplify the elective selection process for students at St. Edward's University. Utilizing a conversational AI, the chatbot aids students in finding courses that match their schedules and academic goals by providing detailed course information. The project aimed to improve user satisfaction and reduce the workload of academic advisors.
role summary
As UX Designer, UX Researcher, and Developer, I led the UX design, contributed to UX research and prototyping, and developed the chatbot using HTML, CSS, Python, JavaScript, Flask, and OpenAI’s Assistant API.
challenges
Identifying Gaps: Currently, St. Edward's University lacks a systematic approach for students to explore and select elective courses that complement their major and minor studies, resulting in potential gaps in both academic and personal growth.
goals
Enhanced Discovery & Autonomy: Our primary goal is to empower students to independently discover and select electives aligned with their personal interests and career goals, while enabling academic advisors to manage their advising workloads more efficiently.
outcomes
Streamlined Processes & Personalized Education: Implementing this solution will streamline elective discovery, boost student satisfaction by offering tailored course recommendations, and reduce the administrative load on academic advisors.
challenges
Identifying Gaps: Currently, St. Edward's University lacks a systematic approach for students to explore and select elective courses that complement their major and minor studies, resulting in potential gaps in both academic and personal growth.
goals
Enhanced Discovery & Autonomy: Our primary goal is to empower students to independently discover and select electives aligned with their personal interests and career goals, while enabling academic advisors to manage their advising workloads more efficiently.
outcomes
Streamlined Processes & Personalized Education: Implementing this solution will streamline elective discovery, boost student satisfaction by offering tailored course recommendations, and reduce the administrative load on academic advisors.
defining stakeholders
Our audience and stakeholders for this website include A Spacious Place, the organization's students, the student's caretakers, organization volunteers
defining stakeholders
Our audience and stakeholders for this website include A Spacious Place, the organization's students, the student's caretakers, organization volunteers
Alex
As Alex, a busy student with a complex schedule, I need help finding electives that fit my availability and career goals so I can enrich my degree plan while balancing my commitments.
challenges
Challenges: Limited availability makes it difficult to find electives that match degree plans without additional guidance.
goals
Goals: Needs elective options that fit specific time constraints and enhance career-related skills.
Alex
As Alex, a busy student with a complex schedule, I need help finding electives that fit my availability and career goals so I can enrich my degree plan while balancing my commitments.
challenges
Challenges: Limited availability makes it difficult to find electives that match degree plans without additional guidance.
goals
Goals: Needs elective options that fit specific time constraints and enhance career-related skills.
Jen
As Jen, an academic advisor managing multiple students, I need quick access to relevant elective options to better support students with diverse scheduling needs.
challenges
Challenges: High demand for personalized advising in addition to regular teaching responsibilities.
goals
Goals: Streamline advising to focus on matching students with relevant courses.
Jen
As Jen, an academic advisor managing multiple students, I need quick access to relevant elective options to better support students with diverse scheduling needs.
challenges
Challenges: High demand for personalized advising in addition to regular teaching responsibilities.
goals
Goals: Streamline advising to focus on matching students with relevant courses.
Stanley
As Stanley, an aspiring entrepreneur and student interested in business, I need an efficient way to discover electives related to business ownership to see if this career path is right for me.
challenges
Lacks time and resources to explore electives fully, depending heavily on advisor support.
goals
Needs efficient access to elective courses focused on business ownership and entrepreneurship.
Stanley
As Stanley, an aspiring entrepreneur and student interested in business, I need an efficient way to discover electives related to business ownership to see if this career path is right for me.
challenges
Lacks time and resources to explore electives fully, depending heavily on advisor support.
goals
Needs efficient access to elective courses focused on business ownership and entrepreneurship.
University Administration
As the central business stakeholder, I need an efficient course selection system to improve student satisfaction and retention, ensuring academic advisors can focus on higher-value tasks.
challenges
Challenges: Limited resources and reliance on advisors for elective recommendations.
goals
Goals: Enhance the advising process to support student success and retention.
University Administration
As the central business stakeholder, I need an efficient course selection system to improve student satisfaction and retention, ensuring academic advisors can focus on higher-value tasks.
challenges
Challenges: Limited resources and reliance on advisors for elective recommendations.
goals
Goals: Enhance the advising process to support student success and retention.
Discovery Process
Our discovery journey began with a problem analysis through the UX Canvas method leading to illustrating our stakeholders, brainstorming solutions, building understanding about development pathways and the market, mapping and testing the solution.
Discovery Process
Our discovery journey began with a problem analysis through the UX Canvas method leading to illustrating our stakeholders, brainstorming solutions, building understanding about development pathways and the market, mapping and testing the solution.
Lean UX Canvas + Proto Personas
We initiated our research with a Lean UX Canvas workshop, collaborating across the class to identify the most pressing issues within the existing advising system. This canvas helped map out the main pain points, risks, goals, and opportunities, including the growing workload of academic advisors and the limited autonomy students have when selecting electives. To deepen this analysis, we also developed proto personas representing key stakeholders like academic advisors, non-traditional students, and those balancing complex schedules. This method provided clarity on the needs of each persona and the challenges they face, giving us a solid foundation for ideation.
Lean UX Canvas + Proto Personas
We initiated our research with a Lean UX Canvas workshop, collaborating across the class to identify the most pressing issues within the existing advising system. This canvas helped map out the main pain points, risks, goals, and opportunities, including the growing workload of academic advisors and the limited autonomy students have when selecting electives. To deepen this analysis, we also developed proto personas representing key stakeholders like academic advisors, non-traditional students, and those balancing complex schedules. This method provided clarity on the needs of each persona and the challenges they face, giving us a solid foundation for ideation.
Landscape and Tool Analysis
After identifying the problem areas, our next step was a Landscape and Tool Analysis. We explored the current tools and systems available for academic advising and elective course discovery, assessing where they fell short in meeting the needs of students and advisors. We reviewed how technology could be leveraged to create a more seamless advising process and looked at competitive solutions in the education space. This analysis highlighted gaps in personalized recommendations, integration of real-time course data, and advisor-to-student communication tools.
Landscape and Tool Analysis
After identifying the problem areas, our next step was a Landscape and Tool Analysis. We explored the current tools and systems available for academic advising and elective course discovery, assessing where they fell short in meeting the needs of students and advisors. We reviewed how technology could be leveraged to create a more seamless advising process and looked at competitive solutions in the education space. This analysis highlighted gaps in personalized recommendations, integration of real-time course data, and advisor-to-student communication tools.
User Journey Flowcharts + Prototyping
With a clearer understanding of the problem landscape, we created detailed user journey flowcharts that mapped out the steps a student would take when interacting with the advising system. These flowcharts captured touchpoints like searching for elective courses, seeking advisor guidance, and finalizing enrollment. Building on these journeys, we developed low-fidelity prototypes to test how our solution could simplify and automate certain parts of the advising process, reducing reliance on manual advisor input.
User Journey Flowcharts + Prototyping
With a clearer understanding of the problem landscape, we created detailed user journey flowcharts that mapped out the steps a student would take when interacting with the advising system. These flowcharts captured touchpoints like searching for elective courses, seeking advisor guidance, and finalizing enrollment. Building on these journeys, we developed low-fidelity prototypes to test how our solution could simplify and automate certain parts of the advising process, reducing reliance on manual advisor input.
Survey and User Testing
To validate our early designs, we conducted a survey and user testing sessions with both students and academic advisors. These tests were critical in understanding how well the solution addressed user needs and pinpointed areas that required refinement. Advisors expressed the need for streamlined course management, while students appreciated the autonomy provided by the chatbot but requested more personalized feedback on their course selections. This feedback was vital in shaping our final iteration, ensuring that the AI-assisted solution was both functional and user-friendly.
Survey and User Testing
To validate our early designs, we conducted a survey and user testing sessions with both students and academic advisors. These tests were critical in understanding how well the solution addressed user needs and pinpointed areas that required refinement. Advisors expressed the need for streamlined course management, while students appreciated the autonomy provided by the chatbot but requested more personalized feedback on their course selections. This feedback was vital in shaping our final iteration, ensuring that the AI-assisted solution was both functional and user-friendly.
Challenges Identified
During the development of the AI-assisted advising chatbot, several challenges emerged that required strategic iteration. These challenges included overlooked details in course-specific information, expanding the chatbot's value beyond basic course recommendations, improving the personalization of responses for unique student profiles, and the need for development team coordination to ensure seamless API integration and chatbot functionality.
Challenges Identified
During the development of the AI-assisted advising chatbot, several challenges emerged that required strategic iteration. These challenges included overlooked details in course-specific information, expanding the chatbot's value beyond basic course recommendations, improving the personalization of responses for unique student profiles, and the need for development team coordination to ensure seamless API integration and chatbot functionality.
Details Overlooked
One of the early challenges was ensuring the chatbot provided accurate and detailed course information. Initial tests revealed that some course-specific details, such as prerequisites and schedule nuances, were being overlooked, which could lead to student frustration if not addressed.
Details Overlooked
One of the early challenges was ensuring the chatbot provided accurate and detailed course information. Initial tests revealed that some course-specific details, such as prerequisites and schedule nuances, were being overlooked, which could lead to student frustration if not addressed.
Expanding Value
During user testing, feedback from upperclassmen students revealed that, while they saw potential in the chatbot, it would be most beneficial for underclassmen who were newer to the elective selection process. Many upperclassmen had already established their own methods for choosing electives and didn’t perceive significant additional value from the chatbot’s current features. This insight highlighted a need to expand the chatbot’s functionality to appeal to a broader range of students by incorporating features like career-aligned recommendations and elective comparisons across semesters, which could also benefit upperclassmen.
Expanding Value
During user testing, feedback from upperclassmen students revealed that, while they saw potential in the chatbot, it would be most beneficial for underclassmen who were newer to the elective selection process. Many upperclassmen had already established their own methods for choosing electives and didn’t perceive significant additional value from the chatbot’s current features. This insight highlighted a need to expand the chatbot’s functionality to appeal to a broader range of students by incorporating features like career-aligned recommendations and elective comparisons across semesters, which could also benefit upperclassmen.
More Personalized Responses
A critical challenge was improving the chatbot's ability to provide personalized responses based on each student’s unique academic needs, schedules, and career goals. Early versions of the chatbot offered generic responses, which led to lower engagement. To address this, we worked on refining the chatbot’s decision-making process by incorporating more personalized inputs from students' profiles.
More Personalized Responses
A critical challenge was improving the chatbot's ability to provide personalized responses based on each student’s unique academic needs, schedules, and career goals. Early versions of the chatbot offered generic responses, which led to lower engagement. To address this, we worked on refining the chatbot’s decision-making process by incorporating more personalized inputs from students' profiles.
Stretching into Development
One of the most technical challenges was ensuring that the development team could effectively implement the AI chatbot’s logic using APIs. While the initial design was strong, it became clear that stretching into the development phase required seamless coordination between UX designers and developers, especially for the integration of OpenAI's Assistants API with the backend systems like Flask.
Stretching into Development
One of the most technical challenges was ensuring that the development team could effectively implement the AI chatbot’s logic using APIs. While the initial design was strong, it became clear that stretching into the development phase required seamless coordination between UX designers and developers, especially for the integration of OpenAI's Assistants API with the backend systems like Flask.
Insights for Iterations
Throughout the development of Topper AI, several key insights emerged, guiding our iterative process to refine the chatbot’s features and enhance its effectiveness. Each insight addressed specific user needs or system improvements, resulting in more comprehensive details, expanded response value, stronger brand alignment, and technical development adjustments.
Insights for Iterations
Throughout the development of Topper AI, several key insights emerged, guiding our iterative process to refine the chatbot’s features and enhance its effectiveness. Each insight addressed specific user needs or system improvements, resulting in more comprehensive details, expanded response value, stronger brand alignment, and technical development adjustments.
Details Overlooked to Details Included
Initial testing revealed that users were frustrated by the lack of specific course details, such as prerequisites, credit hours, and scheduling nuances. To address this, we iteratively updated the chatbot’s information database to include these essential details, allowing students to make more informed decisions when selecting electives.
Details Overlooked to Details Included
Initial testing revealed that users were frustrated by the lack of specific course details, such as prerequisites, credit hours, and scheduling nuances. To address this, we iteratively updated the chatbot’s information database to include these essential details, allowing students to make more informed decisions when selecting electives.
Value Expanded to Offer More Personal and Dimensional Responses
Feedback showed that students desired responses that aligned more closely with their personal goals and career paths. To make the AI more responsive to individual needs, we expanded the chatbot’s response capabilities to include questions about students' aspirations and interests, providing tailored course recommendations that felt more personal and valuable.
Value Expanded to Offer More Personal and Dimensional Responses
Feedback showed that students desired responses that aligned more closely with their personal goals and career paths. To make the AI more responsive to individual needs, we expanded the chatbot’s response capabilities to include questions about students' aspirations and interests, providing tailored course recommendations that felt more personal and valuable.
Brand and Identity Alignment
To ensure Topper AI aligned with St. Edward’s University’s brand and tone, we iterated on the chatbot’s language, making it more conversational, supportive, and aligned with the university’s identity. This branding update helped the AI feel more like an integrated part of the university experience, creating a more cohesive and trustworthy interaction for students.
Brand and Identity Alignment
To ensure Topper AI aligned with St. Edward’s University’s brand and tone, we iterated on the chatbot’s language, making it more conversational, supportive, and aligned with the university’s identity. This branding update helped the AI feel more like an integrated part of the university experience, creating a more cohesive and trustworthy interaction for students.
Development Iterations
During implementation, technical challenges arose, particularly with integrating OpenAI's Assistant API and ensuring data consistency across platforms. To address this, we made several iterative development updates, including optimizing API calls and improving database synchronization, which enhanced the chatbot’s response speed and reliability.
Development Iterations
During implementation, technical challenges arose, particularly with integrating OpenAI's Assistant API and ensuring data consistency across platforms. To address this, we made several iterative development updates, including optimizing API calls and improving database synchronization, which enhanced the chatbot’s response speed and reliability.
Product Features
The Topper AI Advisor offers a suite of features designed to help students find electives that align with their interests, schedules, and career goals. Through personalized recommendations, interactive learning opportunities, and integration with the latest AI technology, the chatbot improves the course selection process for students while reducing the load on academic advisors.
Product Features
The Topper AI Advisor offers a suite of features designed to help students find electives that align with their interests, schedules, and career goals. Through personalized recommendations, interactive learning opportunities, and integration with the latest AI technology, the chatbot improves the course selection process for students while reducing the load on academic advisors.
Find Personalized Electives
Using AI to analyze students' individual preferences, interests, and schedules, the Topper Advisor provides tailored course recommendations that align with each student’s academic and career goals. By incorporating factors such as learning style and desired skills, the system ensures that students are matched with electives that fit their unique profiles.
Find Personalized Electives
Using AI to analyze students' individual preferences, interests, and schedules, the Topper Advisor provides tailored course recommendations that align with each student’s academic and career goals. By incorporating factors such as learning style and desired skills, the system ensures that students are matched with electives that fit their unique profiles.
Interactive Learning
The chatbot offers interactive learning elements that help students explore electives in a hands-on way. Through Q&A sessions and recommendations based on student feedback, the AI enables users to refine their choices and learn more about available electives, creating an engaging, immersive experience.
Interactive Learning
The chatbot offers interactive learning elements that help students explore electives in a hands-on way. Through Q&A sessions and recommendations based on student feedback, the AI enables users to refine their choices and learn more about available electives, creating an engaging, immersive experience.
Latest Technology for AI-Driven Personalization
Topper Advisor integrates the latest advancements in AI to deliver accurate and relevant course recommendations. For example, the system analyzes course descriptions for subtle cues and patterns, using this data to suggest electives that might otherwise be overlooked. This innovative approach provides students with options that fit their personalities, learning habits, and career paths.
Latest Technology for AI-Driven Personalization
Topper Advisor integrates the latest advancements in AI to deliver accurate and relevant course recommendations. For example, the system analyzes course descriptions for subtle cues and patterns, using this data to suggest electives that might otherwise be overlooked. This innovative approach provides students with options that fit their personalities, learning habits, and career paths.
Outcomes and Expectations
These outcome hypotheses outline the anticipated impacts of Topper Advisor on students and academic advisors. While the system has yet to be fully implemented at the university, we predict these outcomes based on testing, user feedback, and advisor validation.
Outcomes and Expectations
These outcome hypotheses outline the anticipated impacts of Topper Advisor on students and academic advisors. While the system has yet to be fully implemented at the university, we predict these outcomes based on testing, user feedback, and advisor validation.
Reduced Bandwidth Load for Advisors
We hypothesize that Topper Advisor will decrease the advising workload by providing students with AI-driven elective recommendations tailored to their preferences and schedules. This reduction in advisor bandwidth will allow academic staff to focus on more complex advising tasks, increasing overall efficiency in supporting students’ academic journeys.
Increased Student Satisfaction in Courses
By matching students with electives that align with their individual goals, interests, and learning preferences, we anticipate higher satisfaction and engagement levels in selected courses. Personalized recommendations should enhance students’ overall academic experience, leading to increased course completion rates and positive feedback about their elective choices.
Opening Doors for Department and Program Exploration
Topper Advisor’s implementation in collaboration with the UX Design Program at St. Edward's University represents a pioneering step into AI-driven education. As one of the university's first student teams to collaborate within an OpenAI TeamSpace, this project sets a precedent, opening up possibilities for other students and departments to explore AI technology in their fields. By showcasing the potential of AI-driven solutions in advising, Topper Advisor may inspire expanded exploration and innovation within the AI realm, fostering interdisciplinary projects and elevating the university's position in the field of educational technology.
Reduced Bandwidth Load for Advisors
We hypothesize that Topper Advisor will decrease the advising workload by providing students with AI-driven elective recommendations tailored to their preferences and schedules. This reduction in advisor bandwidth will allow academic staff to focus on more complex advising tasks, increasing overall efficiency in supporting students’ academic journeys.
Increased Student Satisfaction in Courses
By matching students with electives that align with their individual goals, interests, and learning preferences, we anticipate higher satisfaction and engagement levels in selected courses. Personalized recommendations should enhance students’ overall academic experience, leading to increased course completion rates and positive feedback about their elective choices.
Opening Doors for Department and Program Exploration
Topper Advisor’s implementation in collaboration with the UX Design Program at St. Edward's University represents a pioneering step into AI-driven education. As one of the university's first student teams to collaborate within an OpenAI TeamSpace, this project sets a precedent, opening up possibilities for other students and departments to explore AI technology in their fields. By showcasing the potential of AI-driven solutions in advising, Topper Advisor may inspire expanded exploration and innovation within the AI realm, fostering interdisciplinary projects and elevating the university's position in the field of educational technology.
Project Status + Priorities
This project was part of a senior capstone project and has concluded. All files and insights have been passed on to relevant departments at St. Edward's University and is undergoing further review and analysis by current students in the User Experience Program to evaluate possible implementation.
Project Status + Priorities
This project was part of a senior capstone project and has concluded. All files and insights have been passed on to relevant departments at St. Edward's University and is undergoing further review and analysis by current students in the User Experience Program to evaluate possible implementation.
high priority
Refined User Interaction: Introduce refined threading for outputs that improve readability, avoiding blocks of text. Include visual cues like "typing..." to clear up timing confusion during API interactions.
Technical and Content Enrichment: Continue enhancing the technical backbone and deepen the course-related content to bolster user comprehension and interaction quality.
medium priority
Enhanced Personalization: Elevate Topper’s interaction quality with personalized responses and intuitive focus area groupings. This involves integrating high-value features identified during user testing, like step-by-step registration guidance from the Figma prototype, enhancing usability for freshmen and transfer students.
Additional User Research: Conduct further research to refine focus areas and ensure alignment with user needs, improving the overall effectiveness of Topper.
low priority
Feature Reevaluation: Assess the practical value of elements such as Topper's avatar and certain interface features. Determine whether to maintain, modify, or remove based on their impact on user engagement. Reassess features like Topper’s avatar and certain interface elements for their actual value to user engagement.
high priority
Refined User Interaction: Introduce refined threading for outputs that improve readability, avoiding blocks of text. Include visual cues like "typing..." to clear up timing confusion during API interactions.
Technical and Content Enrichment: Continue enhancing the technical backbone and deepen the course-related content to bolster user comprehension and interaction quality.
medium priority
Enhanced Personalization: Elevate Topper’s interaction quality with personalized responses and intuitive focus area groupings. This involves integrating high-value features identified during user testing, like step-by-step registration guidance from the Figma prototype, enhancing usability for freshmen and transfer students.
Additional User Research: Conduct further research to refine focus areas and ensure alignment with user needs, improving the overall effectiveness of Topper.
low priority
Feature Reevaluation: Assess the practical value of elements such as Topper's avatar and certain interface features. Determine whether to maintain, modify, or remove based on their impact on user engagement. Reassess features like Topper’s avatar and certain interface elements for their actual value to user engagement.
Reflection + Takeaways
reflect
In this project, I ventured beyond my comfort zone, harnessing multi-disciplinary skills in UX and AI to transform a standard course catalog into a dynamic, conversational interface. This not only met a practical need but also fused student needs with institutional goals, illustrating the potential of technology to foster empathetic, personalized user experiences.
learn
Throughout the project, I enhanced my skills across the full spectrum of UX design, including front-end and back-end development, and API integration. I leveraged emerging technologies to craft personalized experiences, solidifying my confidence and focus on pursuing UX and AI in my future career.
reflect
In this project, I ventured beyond my comfort zone, harnessing multi-disciplinary skills in UX and AI to transform a standard course catalog into a dynamic, conversational interface. This not only met a practical need but also fused student needs with institutional goals, illustrating the potential of technology to foster empathetic, personalized user experiences.
learn
Throughout the project, I enhanced my skills across the full spectrum of UX design, including front-end and back-end development, and API integration. I leveraged emerging technologies to craft personalized experiences, solidifying my confidence and focus on pursuing UX and AI in my future career.