Every Bit: A Personalized Movement Journal App

Challenge / Context

I wanted to build a small personal movement-tracking app for myself, but I also wanted to be able to share it with friends and family — who would have very different exercise goals, starting points, and available time than I do. That meant moving beyond a single-user tool built around my own habits, toward something that could genuinely adapt to whoever was using it.

Approach

Try Every Bit — a movement journal that asks each user to identify their own goal, weekly activity level, preferred activities, and comfortable session length, then tailors its suggestions accordingly. I built it using Codex, an AI coding assistant, prompting it specifically to add per-user goal-setting so the app could genuinely serve people with different needs rather than assuming everyone shared mine.

Design Rationale

This mirrors a principle I apply in the classroom — differentiating based on the individual learner rather than a one-size-fits-all default — applied here to personal wellness instead of instruction. Just as importantly, before sharing this tool with anyone, I made a point of verifying exactly how its recommendations are generated. It turned out the app uses a curated set of routines and rules informed by NHS beginner-running guidance, rather than live suggestions generated by an LLM at the moment someone uses it. That distinction mattered to me: predictable, vetted logic is a safer foundation for health-adjacent advice than open-ended AI generation, even though AI tools were genuinely useful in helping me build the app itself.

Outcome / Reflection

This was a low-stakes way to build genuine comfort with AI coding tools outside a classroom context, while practicing a habit I think matters for any instructional designer working with AI tools: not just accepting a tool’s output, but asking where it actually comes from before trusting or sharing it.