“AI-powered coaching” usually means a black box: data goes in, a recommendation comes out, and you're asked to trust it on faith. Ours is the opposite — a short list of rules you can read in one sitting and check against every suggestion it makes.
What the engine sees
The input is simple. When a client logs a session, GainSync has two things: what they actually did against your prescription — sets, reps, weight — and how it felt, rated on a five-point scale from “way too easy” to “way too hard.”
The four rules
Rule one: if the client hit the top of the rep range on every set and didn't call it hard, the engine suggests +2.5% on the bar. Not a hero jump — a small, boring, sustainable increase, because that's the kind that compounds.
Rule two: if the client fell below the bottom of the rep range on any set, it suggests pulling the weight back 5%. Same sets, same rep targets — just a load they can win again next time.
Rule three: if the client missed rep targets and rated the session “way too hard,” it suggests a deload — less weight and roughly half the sets, about 45% less volume. Struggling on both numbers and feel is a recovery signal, not a motivation problem.
Rule four — the one most tools skip: when the signals conflict, the engine does nothing. Hit every rep target but found it brutal? Missed the range but called it easy? Contradictory data produces no automatic suggestion at all. It stays in your review queue, because ambiguity is exactly where coaching judgment belongs.
You stay in charge
Every suggestion arrives as pending until you decide. Accept, and the client's next session updates with the new numbers. Reject, and nothing changes — and the rejection is recorded too, so the history of what you chose stays visible.
No neural network, no mystery. Two inputs, four rules, one person in charge: you.
Run your coaching work in one place
Plans, food, check-ins, progress, and payments — all in one simple app.