Kryvio

Kryvio connects nutrition, training, recovery, health data, and body check-ins to prepare your day, adapt your plan, and explain the next best action.

Comment

Inspiration

Kryvio started with a simple question:

Why do fitness apps still feel like dashboards instead of intelligent companions?

Most apps collect data but leave the user alone to interpret dozens of charts, metrics, and notifications. Nutrition, training, recovery, sleep, and health data often exist in separate places.

I wanted to build something different.

Instead of showing more information, Kryvio focuses on making one meaningful decision at a time.

As a high school student, I wanted to prove that modern AI can help build software that feels more personal, understandable, and genuinely useful instead of simply adding another chatbot to an app.

What it does

Kryvio is an adaptive fitness system that combines:

  • Nutrition tracking
  • AI Food Scan
  • Barcode scanning
  • Workout planning
  • Recovery tracking
  • Body check-ins
  • Health Connect integration
  • Supplement routines
  • Daily briefings
  • An adaptive coach

The goal is not simply to collect data.

Kryvio connects information from different areas to prepare one clear recommendation for the day.

Instead of asking users to interpret everything themselves, the system prioritizes what matters most today.

How we built it

Kryvio is built as a modern cross-platform application using:

  • React
  • TypeScript
  • Capacitor
  • Supabase
  • PostgreSQL
  • Health Connect
  • AI-powered food recognition
  • Codex throughout development

Rather than using AI to generate isolated pieces of code, Codex was used as an engineering partner.

Every major system was reviewed, redesigned, tested, and continuously improved through multiple implementation passes covering architecture, security, accessibility, UI, navigation, testing, localization, recovery logic, and product design.

Challenges we ran into

The biggest challenge was not building features.

It was making everything work together consistently.

Some examples included:

  • keeping nutrition, recovery, workouts, and recommendations synchronized
  • building a trustworthy product instead of making unrealistic AI claims
  • reducing information overload
  • creating a scalable design system
  • supporting localization
  • handling offline synchronization safely
  • improving mobile UX while maintaining performance
  • making recommendations explainable rather than mysterious

Another challenge was balancing ambitious ideas with realistic implementation while still being a student.

Accomplishments that we're proud of

Some achievements include:

  • A unified product architecture connecting nutrition, recovery, training, and health data
  • AI-assisted meal recognition with editable review before saving
  • Adaptive daily recommendations
  • Body check-in system connected to workout planning
  • Recovery Intelligence
  • Secure account isolation and offline synchronization
  • Cross-platform architecture
  • Hundreds of automated tests
  • Strong focus on accessibility, localization, and product quality

Most importantly, Kryvio is evolving from a collection of fitness tools into a connected performance system.

What we learned

One of the biggest lessons was that great software is not created by adding more features.

It is created by reducing complexity.

We also learned that AI should not replace product thinking.

Instead, AI should help users make better decisions while remaining transparent about uncertainty.

Working with Codex taught us how important architecture, testing, documentation, and iterative improvement are for building reliable software.

What's next for Kryvio

Kryvio is still under active development.

Upcoming milestones include:

  • Adaptive Training
  • Advanced Recovery Intelligence
  • Weekly and Monthly Reviews
  • Ultra Concierge
  • Personalized Morning Briefings
  • Smarter cross-domain recommendations
  • Better long-term analytics
  • Expanded Health integrations
  • Improved AI coaching
  • Public beta after extensive testing

The long-term vision is to build a fitness platform that feels less like a tracker and more like a personal performance operating system that prepares, adapts, and explains each day. Kryvio was built by a 16-year-old student from Austria using Codex as an engineering partner, with a parent or guardian acting as the eligible entrant under the official hackathon rules.

Built With

Share this project:

Updates

posted an update —

Major Kryvio Development Update Kryvio has reached a major technical milestone. Over the last development cycle, I focused heavily on reliability, performance, accessibility, AI intelligence, training adaptation, nutrition, recovery, offline behavior, and Android support. Current verification includes: 1,287 automated tests passing Full TypeScript and ESLint validation 0 known dependency vulnerabilities Production and Preview web builds passing Mobile, Capacitor and Android builds passing Responsive QA across multiple phone, tablet and desktop sizes English, German and Turkish localization verified Offline/reconnect and exactly-once synchronization architecture Server-authoritative training progression and reversible recommendations Personalization and bounded outcome learning Body composition and adaptive training systems Recovery, nutrition, AI Coach and Product Truth integration Accessibility improvements including focus handling, large-text support and reduced-motion behavior Kryvio is now entering its final physical Android acceptance phase, where the complete experience is being tested on a real Samsung device, including camera flows, Health Connect, offline recovery, accessibility and runtime stability. The goal remains the same: build Kryvio as a connected performance system rather than another collection of fitness dashboards. More updates soon.

Log in or sign up for Devpost to join the conversation.