Coaching that
can see the set.
Tensio is a coaching platform built on two sensor pods your clients wear while they lift. You prescribe the session. The pods measure every rep on each side: how fast it moved, how long it was under load, and which side gave out first. Your athlete hears the plan at the rack, and you see what they actually did, without standing next to them.
- Athlete
- What’s next?
- Tensio
- Bench press. Your coach has you at two twenty five for eight.
- Athlete
- How did that go?
- Tensio
- Speed dropped eighteen percent by rep eight and the left side slowed first. That goes to your coach with the set.
Kits are not on sale right now. Coaches can pilot now; the list hears first when the next run opens.
$21,000 pre-sold to 60+ Kickstarter backers, a $10,000 grant from the Kickstarter and Google Next Wave Fund, and 10,000 lifters on the list. Prototypes are in daily use. The app already scores every set with a model trained on our own data: 98.2% set accuracy, 99.75% per rep on an 800-set training corpus across fifty exercises. Performance on lifters the model has never seen is the next thing we establish.
See the set your athlete did, not the one they logged.
A remote coach works from what a client types in afterwards. A workout log gives you the result, but little detail about how it was performed. Tensio sends the actual set: every rep, both sides, the tempo and the load, as it happens. You prescribe the session, Tensio records how it went, and you adjust the next one from what happened instead of what was reported.
We are running small paid pilots now: if you coach people for a living and want to see your clients lift, email me about joining the program.
Two pods, because one only sees half.
A single sensor measures one arm and reports it as the set. Your two sides do not move identically, so that number is an assumption about the side it cannot see, and asymmetry is not even available to it: there is no second measurement to compare against. That gap is where a lot of lifters quietly lose progress and build imbalances. Tensio reads two pods at a hundred samples a second and measures each side against the other.
Worn on both wrists. For barbell and dumbbell work the hands are on the load, so what the pods see is what the weight did. Other mounts come once the model has been trained on them, not before.
What you get after a set
- Reps
- Counted per side as you perform them, including the slow ones at the end of the set.
- Tempo
- The concentric and the eccentric of every rep, separately, in seconds.
- Time under tension
- Real seconds of load, with the unrack, the rack and the pauses accounted for.
- Asymmetry
- Left against right, rep by rep, so you watch it develop instead of guessing at it.
The pods are the input layer, not the product.
The variables that actually govern how you adapt cannot be captured by hand by anyone. Nobody times a 1.4 second eccentric while they are under the bar. Nobody holds thirty per-rep durations in their head. Nobody notices their left side began giving out two reps before their right. That is not a discipline problem and it is not something a better logging app solves: those measurements do not exist without instrumentation on the body.
Tensio produces them as a by-product of a normal session, already labeled with the movement and the load because you said both out loud when you started the set. The result is clean, hundred-hertz motion data attached to a known exercise at a known weight, per side, per rep. Nobody has that at scale, because nobody else built a way to collect it that a person will willingly use twice.
That is what makes real personalization possible. Software generalizes: a program in an app is written for a population, and it cannot know that your bench slows at rep six while your squat holds to rep ten, or that you recover from volume faster than the template assumes. Measured per rep, per side, every session, it can.
Inputs
Two synchronized pods, a voice agent that captures the exercise and the weight without you typing either, and a training corpus that grows every time somebody uses the product for the reason they bought it. First batch under production.
Intelligence
A neural network already runs in production, segmenting a lift into its phases and counting reps from raw motion more accurately than the hand-written logic it replaced. That is the loop closing: our data trains a model, the model ships, and every set after it improves the next one. The aim is not one all-encompassing network but specialized models composed together, and fatigue is the one we are building now, read from tempo decay rather than asked about.
Coaching
Every coach on the platform prescribes a session, sees how it went, and adjusts the next one. Those decisions, next to the measured sets they were made from, are what the models learn from. The aim is to extend what good coaches do to the people who never had one: change the plan while you are still standing there, add a set because you have more in you, hold the load on a day your own numbers say to. Against your training, measured, rather than a population average.
Where we actually are.
Built and running
- Working prototypes in daily use, capturing at 100Hz
- PCB and enclosure design complete, in final iteration
- Rep counting, tempo, tension and asymmetry, per side
- A trained model in production at 98.2% set accuracy
- A voice agent that runs the set from the band
- iOS app in beta, with a growing corpus behind it
In front of us
- First production run for the Kickstarter cohort
- First paid coach pilots
- Additional mounts, once the model is trained for them
- Fatigue and proximity-to-failure models
- In-set auto-regulation from velocity and tempo decay
Know when they go on sale.
First word when production units go on sale, plus build notes and test data from the bands along the way. Ten thousand lifters are on the list.