Case study — Kynso
The secret to a 6-domain fitness engine was stripping half the UI off the screen.
Venture
Kynso is an early-stage startup connecting strength, cardio, mobility, nutrition, sleep and, for women, the menstrual cycle into a single ecosystem.
Athletes were juggling three different trackers, a notes app, and ChatGPT just to see their progress. As co-founder, I defined the product roadmap, designed the entire iOS app from scratch, built the brand system, and ran marketing—partnering with one engineer to take Kynso from zero into live internal testing.
The problem
The human body is an interdependent system. Fitness apps track everything in silos.
If you lift in a deficit and ate poorly, a typical fitness app still declares you ready to train because it only sees the last workout. In my user interviews, every athlete had hacked together their own fix across separate tools into ChatGPT. I set out to connect those inputs and analyze the real trade-offs between them.
- Signal Every interviewed athlete used two to three apps plus a notepad and AI to manually spot training patterns.
- Constraint One designer and one engineer building a multi-domain engine and sync pipeline before public beta.
- Unknown Whether users burned by bad tracker algorithms would trust a new, multi-factor readiness score.
How it got built
Hover a track
-
- Product design
“I built myself a GPT thread... I dump everything in there”
Kynso adapts to what you’re doing
Apple Health syncYou’re on the list
Cohort 01 · hybrid athletes
Squat volume dropped 14% after yesterday’s 10k run.
Suggested: drop 1 working set.
“Great job! Let’s get ready for your next set of deadlifts.”
315 lb × 5
1:30
Biological sex
Most fitness algorithms default to a static 24-hour male baseline. Kynso calibrates every calculation—from metabolic burn and lifting volume tolerance to weekly load shifts—against your actual physiology. Without this, the math is guessing.
Beta applicant qualification
- Discipline mix ≥ 2 fitness types Strength exercises + distance running
- Active stack Currently juggling 2+ apps Strava + Strong / Hevy
- Friction point Subscription fatigue, or context lost between disciplines
Screening out low-signal testers, to curate an intentional cohort of hybrid athletes.
The impact
I took Kynso from napkin concept to internal iOS testing, delivering a fully integrated multi-domain build.
- 6
- Integrated health inputs
- 2
- Adaptive UI session modes
- 3x
- Faster logging in 5-person user tests
My biggest breakthrough was learning what to remove. By splitting the interface between calm planning and workout modes, I turned a cluttered tracker into a sharp training tool. I made our readiness engine transparent and cycle-aware, replacing black-box guesswork with clear, actionable coaching.