Autopilot, not tracking
The product owns order, timing and prompts while the user confirms the physical action.
On-device native prototype
GymPilot removes decisions from a beginner's first workout.
The product wedge is not another fitness dashboard. It is gym navigation: show one exercise, one safe starting point, one timer and one next move.
StatusNative SwiftUI prototype installed and launched on a paired iPhone.
UserA beginner who does not know what to do next in a gym.
RiskToo many choices turn a training plan into anxiety and abandonment.
ProofWorking Today, Gym Mode, instruction and progression flows.
Current product proof
The home screen answers one question: what do I do now?
The UI deliberately hides program architecture behind a clear 45-minute session, equipment preparation and a single start action.
- StartOne workout and one primary action instead of a plan catalog.
- DuringManual set start and stop, automatic rest timing and the next prompt.
- SafetyConservative weight calibration with lighter defaults when uncertain.
- ProgressActual reps, feedback and safe increments shape the next session.
Product decisions
Complex logic, quiet interface.
The work is useful to founders building guidance products where the system must decide what matters next without pretending to know more than it does.
Rules before AI
Progression starts with explainable conservative rules instead of opaque recommendations.
Native device proof
SwiftUI, local state, haptics, timers and real-device launch shape the product from the beginning.
Honest boundary
This proves product direction, not market demand.
GymPilot is not public on the App Store and does not yet prove retention, paid conversion or trainer adoption. It is shown as an on-device native prototype, not as a launched business.
ProvedFocused native flow and beginner-safe interaction model.
OpenMarket validation, backend scope, release and monetization.
Relevant when clarity is the feature
Building a product that must guide, not overwhelm?
Send the user, the moment they get stuck and the decision the product should make smaller.