Before taking CS247B, I believed I already had a solid understanding of the design thinking process and how to tailor solutions to specific audiences. Through previous HCI courses and project experiences, I was familiar with user research, prototyping, and iterative design. However, this class expanded my perspective in a meaningful way. I now better understand the depth of considerations required when designing specifically for behavior change rather than just usability or engagement. The course emphasized that influencing behavior involves not only identifying a problem and designing a functional solution, but also carefully studying motivations, psychological barriers, and the broader systems that shape individual actions.
One aspect of the class that stood out to me was the structured progression from broad societal themes to a narrowly defined target segment. In our case, this meant starting with the general topic of AI use among students and gradually refining our focus through proto-personas, story maps, assumption testing, and system exploration. These tools helped our team move beyond surface level observations and build a deeper understanding of the specific group whose behavior we were attempting to influence. I found proto-personas and assumption tests particularly valuable because they allowed us to ground our design decisions in explicit hypotheses that we could test and revise. Similarly, creating moodboards and style tiles was one of my favorite parts of the process. I have always enjoyed curating visual inspiration through vision boards and design references, so translating that instinct into a structured design artifact felt both intuitive and energizing.
At the same time, I occasionally felt that certain components of the course were repetitive. Some frameworks and activities overlapped conceptually, which made portions of the process feel slower than necessary, especially given the limited ten week timeline. While I recognize the theoretical value of tools such as system models, grounded theory, and system paths, in practice they felt less essential for our specific project compared to other methods. If I were to redesign my own workflow in the future, I would prioritize targeted user understanding tools and rapid assumption validation over extensive systems mapping unless the problem space clearly demanded it.
As a senior coterminal student in Computer Science with a focus on Human Computer Interaction, CS247B was my sixth HCI course at Stanford. I initially questioned whether the class would add new value to my learning. However, within a few weeks I became genuinely excited about its central theme of designing for behavior change. I have long been interested in social dynamics and the psychology behind everyday decision making, and this course provided a structured way to explore those interests. A key insight that stayed with me was the recognition that many daily behaviors are deeply ingrained and extremely difficult to change. This realization influenced how my team scoped our project. Although intentional AI use among students is a timely and widely discussed topic, we quickly understood that attempting to change the habits of all AI users would be unrealistic. Instead, we focused on students who were already somewhat motivated to use AI more intentionally but lacked consistent support or feedback mechanisms. Narrowing our audience in this way made the problem more tractable within the course timeframe.
Ethical considerations were another meaningful dimension of the class. I have often felt that discussions of ethics in engineering and computer science are insufficiently emphasized, especially as emerging technologies rapidly reshape daily life. Throughout this project, our team engaged deeply with questions of nudging, privacy, inclusive interface design, and well-being. For example, our prototype incorporated a subtle visual indicator intended to encourage users to reflect on their AI usage patterns and set personal intentions. We aimed to design this feature as a supportive nudge rather than a manipulative mechanism by ensuring that it provided feedback without interrupting users’ workflows or limiting their autonomy. Privacy posed a more complex challenge. Because our concept involved integrating with existing AI tools, accessing real time chat data would have raised significant ethical and technical concerns. As a result, we chose to simulate certain data inputs in our prototype to test our core hypothesis while respecting user privacy boundaries. This experience highlighted how ethical constraints can shape not only implementation details but also the feasibility of entire design directions.
Inclusive design was another theme that shaped our interface decisions. Early visual concepts for our product leaned toward a bold and highly stylized aesthetic that resonated with student culture but risked alienating broader audiences. Over time, we refined the visual language to be more neutral and adaptable while still maintaining a distinct identity. This process helped me appreciate the delicate balance between designing for a specific demographic and ensuring broader accessibility. From a well being perspective, our project sought to support reflective engagement with technology rather than passive or compulsive usage patterns. However, I also recognize that any system that tracks behavior or provides performance feedback could potentially create pressure or self comparison among users if not carefully designed.
After completing CS247B, my perspective on product development has evolved significantly. I now feel more confident that the most challenging aspect of building meaningful technology is not the technical execution itself, but rather identifying the right problem, asking the right questions, and testing assumptions thoughtfully. In an era where rapid prototyping and AI assisted development lower the barrier to creating functional software, the true differentiator lies in understanding human needs and behavioral contexts. Looking ahead, I hope to carry this mindset into future coursework and professional experiences. Instead of focusing primarily on the end product, I want to remain deeply engaged in the reasoning behind what I build and why it matters.
If I encounter a similar design challenge in the future, I would invest early effort in clarifying the behavioral mechanisms at play, defining measurable hypotheses, and selecting a narrowly defined user segment. I would also continue integrating visual exploration methods such as moodboards alongside structured research tools to balance creativity with measurable evidence/feedback. Ten years from now, I believe what I will remember most from this class is not a specific framework, but the lasting reminder that meaningful behavior change requires patience, empathy, and deliberate design choices.
