Ciki Zeng
Live Sample · Real Diagnosis

See exactly what JumpOnion analyzes — start to finish, on one page.

First, a quick look inside the app. Then a complete 2Lo (double loop) diagnosis from a real practice session — the same engine that runs on every customer upload. Look through it, then decide if it's worth $49/month to your skater.

Inside the App · A real practice session

Faces mosaicked for privacy. Recorded inside the JumpOnion app during a real practice session.

1
Step 1 · Physical Metrics

Computer vision measures the jump.

Below are the metrics from one real 2Lo upload. Pose estimation extracts 33 body keypoints per frame; biomechanics turns those points into measurements that match what a coach would see — but exact, repeatable, frame-by-frame.

Air Time
0.348s
target ≥ 0.40s for a clean double
Jump Height
14.9cm
estimated from biomechanical model
Takeoff Velocity
1.71m/s
vertical component
Takeoff Angle
46.6°
near the 45° optimal
Time to Tight
0.214s
target < 0.15s for fast rotation
Under-rotation
82.8°
shortfall from a clean double

Not AI-generated. Computed with computer vision + biomechanics. Same numbers every time you upload the same video.

2
Step 2 · AI Diagnosis

Rule engine measures. LLM translates.

A deterministic rule engine compares every metric to clean-jump targets and decides severity. Then Gemini 3 Pro turns that decision into language a parent and a 9-year-old can both act on. The LLM never invents numbers — it's an interpreter, not a judge.

Primary Verdict · Severity 0
Excellent execution.

This jump looks good overall. No major issues were identified.

Also Noted · Two non-blocking risks
underrotation_risklate_tightening

The skater lacks the vertical impulse needed to achieve sufficient air time for a double jump. Combined with a slightly delayed arm pull, the skater runs out of time to complete the rotations in the air, resulting in an under-rotated landing.

The green "excellent" is the main verdict. The red "also noted" is where most tools stop. That's where JumpOnion starts.

3
Step 3 · Evidence

Every claim is traced to a measurement.

Three evidence points support the "also noted" warnings. Not hallucination. Not guessing. If you want to challenge a claim, click through to the metric.

Air time0.348starget ≥ 0.40s
Strong, reliable measurement. Confirms the height shortage.
Estimated jump height14.9 cm
Corroborates the air-time finding. Independent measurement, same conclusion.
Time to tight position0.214starget < 0.15s
Slow to compress the rotation axis. Costs angular velocity.
4
Step 4 · Personalized Training Plan

The diagnosis becomes a sequence.

Four off-ice drills, picked from a 54-drill catalog, ordered by when they should happen relative to the next ice session. Each drill explains why— so you know what you're fixing, not just what to do.

01

Metronome-Synced Takeoff

3 × 8
Pre-Ice Activation

Builds consistent, explosive vertical impulse — train your body to be a precise 'human metronome' for maximum jump height.

02

Kinetic Freeze-Frame

3 × 8
Pre-Ice Activation

By forcing instant 'freeze-frames' in various jump positions, this drill sharpens proprioception so you can find and hold a perfectly tight axis in the air.

03

Fast Arm Pull

3 × 15
Pre-Ice Activation

Targets tightening speed directly. Train the arms to snap in like a retracting tape measure the instant the blade leaves the ice.

04

Band-Resisted Arm Snap

3 × 12
Post-Ice Strengthening

Builds explosive 'braking' power against centrifugal force. Essential for snapping into a super-tight air position when rotation is already loaded.

5
Step 5 · Skater Cue + Observer Radar

What the skater feels. What the parent watches.

Most AI tools forget that the people executing the fix are a 9-year-old on the ice and a parent in the stands. JumpOnion gives each of them their own page.

Skater Cue · Mind picture

The instant your blade leaves the ice, imagine your body becoming a tightly wound spring — push up, then instantly snap into a spinning top.

Observer Radar · What to watch

Stand at the side, slightly behind the takeoff point. Watch the takeoff knee: does it fully extend upwards, like a powerful piston, or is there hesitation? Immediately after takeoff, watch the arms: do they snap to the chest in one explosive motion, or is there a visible delay?

Want this for your skater?

Every JumpOnion subscription includes unlimited jump uploads, full access to the 54-drill library, historical tracking across sessions, before/after comparison, and email support from a skating mom (me).

See Pricing →Starts at $49/month · cancel anytime
Why not just use Gemini directly?

I tested it. Gemini misclassified a 2Lz as a Toe Loop. Gemini called an imperfect air position "perfect". LLMs hallucinate; rule engines don't. JumpOnion uses Gemini only as a translator on top of a deterministic biomechanical engine — that's what makes the diagnosis trustworthy. Read the full breakdown on the blog.

Questions?

DM @cikibuilds on X or @jumponion on 小红书. I read everything.

The diagnosis, metrics, and training plan above come from a real production task (0dc3eb30, March 30, 2026). Not a mockup. Not a demo. The same engine ran on this upload as runs on every customer upload.