2
1 Comment

After 20 years of yoga practice and a career in product, I built the app I couldn't find

After years as a product lead, I kept noticing the same thing: yoga apps are built backwards.

A real class isn't visual. You walk in, the teacher speaks, you move. Apps do the opposite — hundreds of classes, endless browsing, more content than you'll ever need. Intimidating before you even begin.

I've practiced for over 20 years. None of the apps came close to replicating what a real class feels like. So I built one.

Honestly, I couldn't have done this a few years ago. I'm not a developer, and the budget constraints were real. AI changed that equation completely. So when the tools caught up, I stopped waiting.

It wasn't a straight line. One of the hardest pieces was audio — pre-generating pose-level guidance that actually sounds like a teacher, not a robot. ElevenLabs got me closest, though even that had its challenges. It also ended up being the best tool for background music — with its own credit management headaches.

Full article on that here: https://www.yogakosh.com/blog/i-thought-finding-a-yoga-voice-would-be-easy-it-rebuilt-my-entire-app

The other major challenge was illustrated pose images at scale — in a consistent style I actually liked. A graphic designer was never in the budget. A combo of DALL-E and Nano Banana got me there.

Article on that process here: https://www.yogakosh.com/blog/from-1-to-100-automating-asset-creation-for-my-yoga-app

The result is yogakosh — audio-first, no catalog to browse, a flow ready when you open it. Or build your own in three taps. Then put the phone down.

It launched on the App Store today.
appstore link: https://apps.apple.com/us/app/yogakosh/id6761315127

If you practice yoga, I'd genuinely love feedback. If you're a builder, I'm curious what questions you have about going solo on something like this.

Product hunt link: https://www.producthunt.com/products/yogakosh?launch=yogakosh
website: https://www.yogakosh.com/

on June 2, 2026
  1. 1

    This is a strong wedge because you’re not trying to build “another yoga app.” The audio-first decision changes the whole category frame.

    Most yoga apps make the user browse, choose, compare, and watch. Yogakosh seems closer to recreating the feeling of walking into a class where the teacher already knows how to guide the session.

    That positioning is worth leaning into harder. The promise is not just “AI yoga flows” or “personalized yoga.” It is more like: open the app, hear the class, put the phone down.

    For launch, I’d focus less on general productivity/wellness users and more on people who already practice but hate screen-led fitness apps: busy professionals, older practitioners, people returning to yoga, and anyone who wants guidance without staring at a phone.

    The story is strong. I’d make the launch message more about “yoga without the screen dependency” than the AI build process.

Trending on Indie Hackers
Your build-in-public audience is not your market. I learned the difference the slow way. User Avatar 122 comments I built a WhatsApp AI bot for doctors in Peru — launched 3 weeks ago, 0 paying customers, and stuck waiting for Meta to approve my app User Avatar 60 comments From broke and burned out as a PM, to launching my SaaS and optimizing my health User Avatar 30 comments Built a "stocks as football cards" thing. 5 days in, my launch tweet got 7 views. What am I missing? User Avatar 26 comments I kept starting projects and dropping them. So I built a system that wouldn’t let me User Avatar 23 comments We built Shopify themes to $20k/month. Now we have to pivot. User Avatar 22 comments