Skip to content
Lia
Industries

Everyone downloads it in January.

iLeaf has built AI-personalised fitness applications, custom training products and platforms connecting athletes to trainers. The engineering problem in this sector is not tracking but retention: the data is easy to collect and almost every app loses its users within weeks, so personalisation and progression are the product rather than features on top of it.

Talk to a solutions lead See client results
Ask Lia about sports & fitness

What this includes

AI-personalised programmes
Plans that adapt to what someone actually completed rather than what they intended in week one.
Coach & trainer platforms
Connecting athletes to trainers, with programme assignment and progress visible to both.
Wearable & sensor integration
HealthKit, Google Fit and device SDKs, reconciled when the same session arrives twice from two sources.
Progress & habit mechanics
Streaks, milestones and progression designed around the drop-off point rather than the launch week.
Video & form guidance
Exercise demonstration and technique feedback, including computer-vision-assisted form checking.
Nutrition & recovery
The rest of the picture, because training data alone explains only part of an outcome.

How we run it

Every phase ships something usable on its own, so you are never holding a half-finished system waiting on the next milestone.

  1. Design for week six

    Launch-week engagement is not a signal. The product is judged on whether someone is still using it after the initial motivation is gone.

  2. Adapt to the completed session

    A plan built on intentions breaks on the first missed week. Personalisation works from what was actually done.

  3. Be careful near health claims

    Fitness sits close to medical advice. The line is drawn explicitly, and the product stays on the training side of it.

  4. One source of truth per session

    Wearables double-report constantly. Deduplication is designed in, because a user who sees a phantom workout stops trusting the numbers.

What we build it with

  • Swift
  • Kotlin
  • HealthKit
  • Google Fit
  • Computer vision
  • Python
  • PostgreSQL
  • Firebase
  • Push / notifications
  • Recommendation models
  • AWS
  • Analytics

Questions we get asked

Have you built fitness applications before?

Three of them: an AI-based personalised fitness solution, a custom fitness mobile application built around a client's own programme, and a platform for sports trainers focused on athlete performance. All three are in the case studies.

How do you handle retention?

By treating week six as the design target rather than launch week. Programmes adapt to sessions actually completed instead of the plan someone set on day one, and progression mechanics are built around the point where motivation drops rather than the point where it is highest.

Can you integrate wearables?

Yes — HealthKit, Google Fit and device SDKs are routine. The part worth planning is reconciliation: the same run frequently arrives from a watch and a phone, and a user who sees a workout counted twice stops trusting every other number in the app.

Related

Let’s talk about sports & fitness.

Tell us what you are running and what it needs to do next. We will tell you honestly whether we are the right team for it.

Talk to a solutions lead
Selected work

Sports & Fitness: the work behind this