Mobile · Health & Fitness

FitSherpa — your personal trainer in your pocket

A mobile app that generates AI-powered workout plans, guides users through every set, and surfaces real progress over time — for people who want to train smart without paying for a personal trainer.

Client chdr.tech (in-house product)
Platforms iOS · Android
Role Design · Development · DevOps
Year 2026
FitSherpa - home screen
Case study

A digital coach that knows your schedule, gear, and goal

Most fitness apps force a trade-off — either a one-size-fits-all plan or live coaching behind a paywall. FitSherpa bridges both: a tailored plan generated from a 60-second onboarding, a guided session with set-by-set logging, and meaningful progress tracking over time.

The challenge

Build a product that, within a minute of opening the app, hands the user a personalized training plan — with no expert knowledge, no surveys, no manual configuration. The plan has to be realistic for the user’s constraints (time, equipment, goal) and evolve with their progress.

The solution

A cross-platform Flutter app backed by Supabase, with workout plans generated by AI through Edge Functions. Four-question onboarding, automatic plan generation, guided sessions with RPE/RIR, full lift & body history, full PL/EN localization, and Google & Apple sign-in.

Key features

Everything a coach would tell you on day one

Each feature answers a real user question: “How much should I lift today?”, “What did I do last week?”, “What if I don’t have any equipment?”. The modules below are those answers.

AI-generated plans

Onboarding captures hours per week, location, equipment and goal — AI picks the exercises, sets, and progression. The plan is ready before the user leaves the onboarding flow.

Guided workout sessions

Step-by-step mode with exercise animations, inline rep and weight inputs, automatic saving, and RPE/RIR tracking for intentional intensity.

Last-lift memory

For every set we surface what the user lifted last time on this exercise — no more guessing whether “today is a weak day”.

Free workout mode

Plans don’t always survive contact with the gym. Free Workout lets users add exercises on the fly in list or guided view — every set is still logged into history.

Progress photos

An optional prompt after each workout to add a body photo. The timeline gallery shows the changes numbers can’t.

Stats & measurements

Per-exercise progress charts, total tonnage, training frequency, body measurement tracking, and a weight history — all in one place.

Apple & Google Sign-In

One-tap auth compliant with both store guidelines. No passwords, full onboarding from first screen to a ready plan.

EN / PL localization

Fully localized UI in two languages — including plan names and workout reminder push notifications.

Smart workout reminders

Local notifications fire only on planned training days, at a time the user picks. They never spam on rest days.

UI

A consistent dark theme designed for the gym floor

Every screen was tested under real gym lighting — high contrast, large tap targets for sweaty hands, and animations that stay subtle and unobtrusive.

Tech stack

Cross-platform, serverless, ready to scale

The architecture was picked so a single team can maintain iOS and Android without compromise, while backend cost only grows once real user traffic does.

Flutter Dart Supabase (Postgres) Supabase Edge Functions Supabase Storage Riverpod go_router OpenAI / LLM Sign in with Apple Google Sign-In Local Notifications flutter_animate App Store Connect Google Play Console

Have a similar product idea?

At chdr.tech we build mobile and web products end-to-end — from the first Figma stroke to live releases on the App Store and Google Play. Let’s talk.

Get in touch with chdr.tech