Build Log 0: msPhysio

I began building msPhysio in the fall of 2025 with the help of AI-assisted development tools. At the time, many of my peers were under significant time pressure, and I was also returning to graduate-level physiology after nearly a decade away from formal health sciences training. I wanted to keep my software engineering skills sharp while building something genuinely useful. What I initially assumed would be a straightforward process of shipping an idea from GitHub to the app store quickly proved otherwise.

So far, I have spent over 100 hours developing the application, and I likely should have tracked that more carefully. While AI tools accelerated parts of the process, the work still required independent problem-solving, debugging, and architectural decision-making. AI can only do what you know how to ask it to do, and when you do not understand the system you are building, that limitation becomes clear very quickly. In many ways, this project reinforced the importance of foundational computer science skills. I am particularly grateful to Oregon State University’s post-baccalaureate computer science program for providing the technical grounding and judgment needed to build responsibly.

I also underestimated how many constraints exist when bringing an educational application to production. App store requirements, platform rules, and content considerations introduced layers of complexity I had not initially anticipated. More importantly, the process deepened my appreciation for educators who design assessments that scale in difficulty and meaningfully test understanding rather than simple memorization.

The natural question is whether the app actually helps.

For me, it does. That said, it is not a shortcut, and it did not magically eliminate the work required to master the material. It certainly did not guarantee perfect grades, and that is not what it is designed to do. msPhysio works best as a supplement, helping reinforce recall and familiarity with a large volume of content, particularly during short study intervals. Like any tool, its effectiveness depends on how it is used.

Because the app has not yet been broadly released, I cannot yet speak definitively to its impact on others. Many of my peers currently rely on tools like Anki and Quizlet, which each address this problem in different ways. Anki offers powerful spaced repetition but has a steeper learning curve and a less intuitive mobile experience for some users. Quizlet made studying more accessible for many students, though its move toward subscription-based access may not suit everyone. msPhysio explores a different balance point, emphasizing immediacy, clarity, and low-friction mobile study.

There are well-resourced academic and industry teams working on similar challenges, often with more advanced AI capabilities. My approach has been intentionally incremental, focused first on serving a local academic community and learning through direct execution before expanding functionality. This build log documents that process transparently, including what I build, how long it takes, what it costs, and the decisions involved along the way.

Ultimately, msPhysio is both an educational tool and a personal experiment in execution, sustainability, and scale. I am documenting the process openly to evaluate whether it can meaningfully support learners and, over time, help offset the significant financial burden of medical education, while also keeping me honest about what it actually takes to build something from scratch.