AI Ayurveda that remembers you — no streaks, no guilt
4.8(6)10StartupsHealthMeditation
Hi Fazier 👋 I'm Ganesh.
Every wellness app I've tried optimises for my engagement — streaks, push-nudges, a little guilt when I skip a day. I wanted the opposite. So we built it.
InnerVeda is Vaidya, an AI Ayurvedic companion that learns your body type and actually remembers you — that you sleep badly when you eat late, that you dislike milk, that Yoga Nidra at 10pm works for you. It builds a personalised 90-day plan for one thing you care about (sleep, stress, digestion, or energy) and adapts it every week.
Three deliberate choices:
• It remembers — the plan gets more personal over time, not more generic. ("How did it know?" is the moment we optimise for, not screen time.)
• No dark patterns — no streaks, no guilt. A 48-hour rule pauses Vaidya when life gets busy; your plan resumes where you left off.
• It can't make things up — every protocol is read from the classical texts (Charaka, Suśruta) and cited; where modern science agrees, we show it.
Start with the free 3-minute body-type quiz — no signup, instant result: 👉 https://innerveda.app/body-type-identification?utm_source=producthunt&utm_medium=referral&utm_campaign=launch_jun2026
Live on iOS, Android & web. Already #1 of the day on Fazier and PeerPush. I'd love your honest take — what did the quiz get right, or wrong? AMA.
4.8
6 ratings
Fractional CTO & CPO | Built $140M+ ARR | AI-first engineering leadership for startups | Kompella Technologies
AI video with native sound, paid per clip. Nothing stored.
A calmer Mac menu bar, one click away.
Enterprise Software. Intelligent Automation. Global Impact.
Comments
10Fractional CTO & CPO | Built $140M+ ARR | AI-first engineering leadership for startups | Kompella Technologies
I want to share the behind-the-scenes — the two problems we lost the most sleep over. 🔬 Making an AI you'd actually trust with your health. "AI + Ayurveda" sets off an obvious alarm: a chatbot confidently inventing remedies. So we built the opposite of a free-text oracle. Your body-type result comes from a deterministic scorer — plain, auditable code, not the model — that a qualified Ayurvedic physician can sign off on. Vaidya is then grounded only on the classical texts (Charaka, Suśruta) and your own data, and constrained to *apply* documented guidance with sources, never improvise it. When it doesn't have a grounded answer, it says so. Teaching the model to be humble was harder than making it smart. 🌱 Building it to graduate you, not keep you. Most apps are designed so you never put them down. InnerVeda's plan ends — on purpose. The 90-day arc has a real finish line, and a 48-hour rule quietly pauses things the moment life gets busy: no guilt, no streaks to "break." It almost certainly costs us the engagement numbers investors like. But the moment a wellness app optimises for screen time over the person, it's lost the only thing that matters. That's our bet: that calm, honest and personal beats sticky. Two things I'd genuinely love from you: - Take the free quiz — does Vaidya's first reading actually feel like it *gets* you? Tell me where it's right or wrong. - If you build AI products: how do you handle trust and hallucination in a high-stakes domain? I'd love to compare notes. Thank you for being here 🙏
Frontend Developer
All this kind of apps are helpfull. Have you thougth about how to monetize them ?
Fractional CTO & CPO | Built $140M+ ARR | AI-first engineering leadership for startups | Kompella Technologies
Good question. We're subscription based, a 90 day personalised plan for one thing (sleep, stress, digestion or energy) that adapts weekly. No ads, no selling data, no upsell funnels inside the app. The body type quiz is free and always will be, that's the trust builder before anyone pays for anything.
Founder of Lapsewise
Love the "no streaks, no guilt" philosophy. That's a refreshing take compared to most wellness apps. Wishing you a great launch! 🚀
Fractional CTO & CPO | Built $140M+ ARR | AI-first engineering leadership for startups | Kompella Technologies
Thank you, means a lot coming from another founder. Honestly the guilt mechanics in most wellness apps are what made me want to quit them, so building the opposite felt personal rather than strategic. Wishing Lapsewise a great run too, will check it out.
What technical solutions does an AI team typically need to adopt to strictly limit AI hallucinations and ensure that it only outputs authoritative and documented content?
The deterministic, physician-signable body-type scorer is the detail that makes the rest credible: pulling the classification out of the model and into auditable code is exactly how you keep trust in a high-stakes domain. We made a similar call in a different field. For anything that has to be defensible we generate by constrained substitution from vetted source material instead of letting a model free-write, so there is simply nothing to hallucinate. And designing the 90-day plan to actually end, with a 48-hour pause instead of streaks, takes real conviction. Took the quiz and the first reading felt specific rather than generic. Congrats on the launch.
Fractional CTO & CPO | Built $140M+ ARR | AI-first engineering leadership for startups | Kompella Technologies
Really appreciate this. You get it exactly right, the scorer being deterministic and auditable code rather than a model output is the whole point. When something touches sleep or health people deserve to know the answer isn't a guess dressed up as confidence. Same logic applies to the AI companion itself, every protocol it gives is retrieved from Charaka and Suśruta, cited, never generated fresh. Would love to hear more about your constrained substitution approach, sounds like we landed on similar principles from different directions. And yes, the 48 hour pause took some internal debate. Growth people hate it. We kept it anyway.
Optimizing for a personalized, adaptive 90-day plan rather than maximizing daily screen time is an excellent approach to digital wellness. Congrats on hitting #1 of the day!
Fractional CTO & CPO | Built $140M+ ARR | AI-first engineering leadership for startups | Kompella Technologies
Thank you, do take it for a spin on mobile. It's really nice.