Private by architecture
The foundation is designed for offline-first sessions. Raw video is used for analysis but is not persisted or uploaded.
MotionLab PerformanceNative iOS foundation
Private, offline-first iOS performance tracking
Track human performance without turning raw video into cloud data.
A native iOS foundation for private movement and training analysis, designed around on-device capture, reviewable metrics, longitudinal sessions, and scientific validation gates.
EvidenceRaw video is not persisted or uploaded. Current production builds remain fail-closed until signed deployment, billing, backend, and reference-comparison evidence are approved.
MotionLab PerformanceLOCAL TRAINING SESSION
ATHLETE / SESSION
Squat session review
iOS 17+ native foundationRaw video is not retainedReference comparison pending
Product overview
What MotionLab Performance is built to do
Clear product boundaries are part of the design: capability, evidence, and current maturity are stated separately.
Performance, not diagnosis
The product organizes training observations and experimental metrics; it does not diagnose injury, prescribe treatment, or replace qualified coaching or clinical judgment.
Release remains fail-closed
Staging and production require explicit environment, backend, commerce, feature, signing, and billing configuration rather than falling back to mock services.
Workflow
From source evidence to a reviewable result
- 01
Capture
Record a supported activity with the device camera using an activity-specific setup.
- 02
Calibrate
Complete the defined camera and scale checks required by the selected metric workflow.
- 03
Review
Inspect pose quality, repetition or cycle boundaries, movement phases, and any experimental metric qualifications.
- 04
Compare
Use local session history, baselines, and repeatability context to examine change over time.
- 05
Export
Create structured scientific or athlete-facing datasets while keeping raw video outside persisted exports.
Capabilities
A focused professional toolset
Each capability supports the product's defined workflow; none is presented as a substitute for qualified review.
Activity analyzers
Implemented foundations cover back-squat, biceps-curl, and treadmill-running workflows with activity-specific metrics.
On-device pose
A Vision-based provider and explicit tracking-quality engine support local analysis and quality disclosure.
Session history
Local sessions, baselines, records, and training views support longitudinal review without requiring cloud video storage.
Experimental velocity
A clearly labelled experimental estimator supports implement-velocity research for defined squat workflows.
Scientific exports
Structured datasets and XLSX-oriented export models retain metric definitions, units, versions, and validation states.
Controlled commercial activation
StoreKit, subscriptions, entitlements, and AI-service boundaries are implemented behind explicit environment and release gates.
Product interface
The work stays connected to its context
Representative views of the implemented native iOS foundation.
MotionLab PerformanceLOCAL TRAINING SESSION
ATHLETE / SESSION
Squat session review
Training session
Capture, pose quality, activity selection, and metric context are organized for a focused mobile workflow.
MotionLab PerformanceLOCAL TRAINING SESSION
ATHLETE / SESSION01ActivityReviewed 02Movement metricLinked 03Baseline comparisonExperimental
Squat session review
Movement replay
Review repetition phases, joint metrics, quality state, and video-derived observations without retaining raw footage.
MotionLab PerformanceLOCAL TRAINING SESSION
ATHLETE / SESSION
Squat session review
SESSION HISTORYStored on device
Raw video is not retained after on-device analysis.
History and baselines
Session records support longitudinal comparison while preserving metric maturity and validation status.
Example outputs
Outputs designed to be reviewed—not merely generated
Local performance record
Activity, timing, movement metrics, quality, and version context without raw video.
Scientific dataset
Structured export tables support defined reference-comparison and research workflows.
Athlete-facing export
A controlled summary communicates observations while preserving experimental and validation labels.
Supported file formats
Inputs and outputs, without ambiguity
PurposeFormatsNotes
CaptureNative iOS camera session
Raw video is analyzed transiently and is not persisted or uploaded.
Local recordsStructured app data
Sessions and baselines remain on the device in the current foundation.
Scientific exportXLSX and structured tabular data
Definitions, units, algorithm versions, and validation states accompany metrics.
DeploymentSigned iOS application
TestFlight and production activation require approved Apple and server configuration.
Privacy and data handling
Know where the evidence goes.
- Raw video is not persisted or uploaded by the current product foundation.
- Development and test builds may seed a local sample session; staging and production builds do not.
- A future AI insight endpoint is designed as a server-mediated service and is disabled without explicit deployment configuration.
- Production configuration rejects server secrets and requires client-safe HTTPS endpoints and public identifiers only.
Validation
Engineering validation infrastructure implemented; scientific reference comparison pendingEvidence has boundaries.
Metric definitions, protocol templates, checksums, dataset validation, and evidence analysis are versioned. Metrics remain experimental or reference-comparison-pending until separately reviewed real evidence supports promotion.
- The validation harness rejects ambiguous units, coordinate conventions, duplicate records, and checksum mismatches.
- Current metric definitions cover squat, curl, and treadmill-running timing, angles, range, cadence, repeatability, and experimental velocity.
- Production use requires independent reference data, explicit approval, signed deployment, and operational configuration.
Pricing and access
Access that matches the product's maturity
No hidden enterprise tier, artificial urgency, or unsupported assurance.
Engineering and validation collaboration
By request
For qualified researchers, performance professionals, and technical partners able to support defined reference comparison.
- Protocol review
- Reference-system alignment
- Evidence and limitation review
Public or TestFlight access
Not available
No public App Store, TestFlight, subscription, or production service is represented as active.
- Signed release still gated
- StoreKit identifiers remain unset
- No production metric claims
Frequently asked questions
Technical questions deserve direct answers
Is MotionLab Performance available in the App Store?
No. The native iOS foundation is implemented, but public App Store and TestFlight access are not represented as active.
Does it upload or keep my workout videos?
The current architecture does not persist or upload raw video. Structured session metrics and local history are separate from the transient capture.
Which activities are currently represented?
The implemented foundations include back-squat, biceps-curl, and treadmill-running workflows. Metric maturity varies and remains explicitly labelled.
Are the metrics scientifically validated?
Not yet as general product claims. The reference-validation infrastructure is implemented, but the current registry remains reference-comparison-pending and experimental by default.
Is this a coaching or medical device?
It is a performance-analysis foundation. It is not presented as an injury-diagnosis, treatment, rehabilitation, or automated coaching authority.
Documentation
Review the method before relying on the output
Metric definitions
Versioned definitions record units, camera view, coordinate conventions, algorithm versions, and required reference alignment.
Reference-validation protocols
Templates define how real comparison datasets should be prepared, checked, and reviewed before a metric can advance.
Release boundary
Production remains fail-closed until signing, billing identifiers, client-safe endpoints, feature gates, and backend operations are explicitly supplied.
Contact
Evaluating an on-device performance workflow?
Share a non-confidential description of your use case, users, and current process. Biosystems One will respond with the appropriate product and access information.