One set becomes a clear plan for the next.
How many more reps you could have done (lifters call this reps in reserve, or RIR) decides almost every weight you pick. It is also the number people are worst at guessing, and the error is not random: when you are tired, most lifters think they are closer to failure than they are. So it is worth measuring rather than asking about.
Same lifter, same load. Rep 1 against rep 5.
Four steps from camera to next set.
The camera watches your joints
A pose model (software that finds your joints in the video) runs on your phone and follows the joints that matter for the lift you picked. Nothing is uploaded: the video stays on your phone and only the numbers made from it are kept.
Each rep is timed, not just counted
A rep is a shape, not a tap. The tracker finds the turnaround at the bottom and the lockout at the top, which gives the time it took to lift (lifters call this the concentric). The change in that time across a set carries the information.
The slowdown is the reading
As a set gets close to failure, even your fastest effort gets slower, and it does so in a steady way for each person. Comparing the last rep against the earlier ones in the same set estimates how many reps you had left.
The next set is written from it
A plan has a target effort, not only a weight. If you had more left than the target, you had room, so the next set gets heavier. If you had less left, the weight holds or drops.
It will not tell you a set was good.
It tells you what it measured and what it could not. If the camera lost the bar, the angle was wrong, or the set was too short to read, it says so instead of guessing. Estimated numbers are shown in a different color from measured ones, everywhere they appear.
Shallow and fast reps still count.
A rep that is not deep enough is flagged, not thrown out: the effort was real. A rep that gets back to the top in under 0.3 seconds is flagged too, counted once, and left out of the slowdown estimate.