NADI | Devpost

What Inspired Us

It didn't start with a brief. It started with exhaustion.

We were sitting together one evening, the kind of evening where you've been "productive" all day but feel completely hollow. Tabs open everywhere. A notes app full of half-formed ideas. A to-do list that had been staring back at us for three days. Somewhere between talking about AI trends, career pressure, and what we actually wanted to do with our lives, one of us said something that stopped the conversation cold: "I have so many ideas. I just never have the energy to do anything about them."

That was it. That was NADI.

We're young adults living through a genuinely overwhelming moment. The AI era has raised expectations and flooded every waking hour with things to learn, catch up on, and act on. And the more we try to keep up, the more we burn out. The more we burn out, the less we actually do. The more the ideas pile up unacted on, forgotten.

We sat down and spoke honestly to each other. Not pitching ideas, questioning things. We criticised every tool we used. Calendars treat the mind like a machine with a schedule. Productivity apps add more decisions to an already overloaded system. None of them asks the question that actually matters: how much do you have left right now?

As the conversation kept going, we landed somewhere unexpected, in ancient Indian frameworks. Nadi, in Sanskrit, means channel: the path through which life force moves. The same tradition describes three energy channels, Ida (rest), Pingala (drive), and Sushumna (balance), and three dimensions of mental energy: Prana (vitality), Tejas (focus), and Ojas (deep reserve).

We weren't building a wellness app. We were translating a 3,000-year-old model of how the mind and body communicate into something a burnt-out young adult could wear on their wrist and feel, not read.

What We Learned

Burnout is a sensing failure, not a productivity failure. We don't crash because we work too much. We crash because we have no instrument telling us we're running low. The insight that changed everything: we didn't need to help people work more; we needed to help them know themselves better.

Ancient frameworks are surprisingly precise. What Ayurveda calls Ojas maps almost exactly onto what neuroscience calls cognitive reserve. Ida and Pingala map onto the balance between the parasympathetic and sympathetic nervous systems. The vocabulary already existed 3,000 years ago. We just needed to listen to it.

Design for the depleted user first. Our most important user is the one least able to engage with a complex interface. When someone is overwhelmed, everything must be one tap, one breath. If it works for them, it works for everyone.

Voice is the most honest interface. When your mind is full, you can still speak. Typing is a second job. The moment we committed to press-and-speak capture, the whole product clarified.

Honest conversation beats every design process. Our best decisions came from the same place, the idea did not pitch to each other, but genuinely questioned each other. Every time one of us said, "I don't think that's right," the product got better.

How We Built It

We split the build into two tracks: the band and the app.

The band communicates through temperature, no screen, no notifications. A warm band means your cognitive energy is strong. A cooling band means depletion is starting. A neutral band means you're steady. The central sensor node, The Seat, sits where all three braided strands converge and simultaneously measures skin temperature, heart rate variability, and skin conductance. There's one physical button press and hold, speak, release. One haptic pulse confirms. The thought is held. Nothing else happens right now.

The app was built in Figma for design and prototyped in React + TypeScript. It has 11 screens connected in a deliberate journey:

Welcome — the invitation, two actions only
Onboarding — three permission cards framed as initiation, not a checklist
Wristband Config — live band preview across all three thermal states; pick your strap colour
Tutorial — three swipeable pages teaching the energy channels, the mental energy system, and the five cognitive states in plain language
Home Dashboard — energy score, channel bars, a live NADI suggestion, and quick access to Capture and Restore
Speak — one button, ripple animation, minimal copy
Captured — calm green confirmation, auto-categorised, suggested time window
Revisit — your thought inbox: voice notes with waveform, transcript, trim, and share; a calendar strip showing today's energy windows
Restore — personalised recovery suggestions with the reasoning behind each one
Profile — settings, data transparency, and the 72-hour sovereignty protocol

Alongside the app, we built a standalone brand site in HTML/CSS/JS — a manifesto, interactive thermal state tabs, a colour picker with a live band preview, and a waitlist.

The Challenges We Faced

Making the invisible legible. Cognitive energy is personal and non-linear. Turning it into something a wristband expresses and a first-time user feels immediately requires more iteration than anything else. The thermal metaphor was the breakthrough: warmth already means capacity in every culture. No translation needed.

Balancing ancient wisdom with modern product sensibility. Sanskrit is precise and beautiful. It's also unfamiliar. We built a layered system: plain English first, Sanskrit as secondary labels during onboarding, then both step back and let the experience speak for itself. By the time someone reaches the home screen, they already understand the concepts in their own words.

Getting the voice capture moment right. Speaking into a device feels vulnerable. The recording screen went through more passes than any other. The final copy "Say anything on your mind. Tasks, ideas, worries, or questions." is longer than every earlier draft and far more human. Every earlier version felt like a command.

Designing for depletion. The Restore screen is the most complex screen in the app and the one that must feel most effortless, because it opens when someone has the least left to give. Rich enough to be useful, simple enough to use while running on empty. That tension was never fully resolved. We kept returning to it.

Scope and time. We had more ideas than hours. Deeper calendar integration, AI-powered resurfacing, and weekly pattern analytics exist in the blueprints but not fully in the prototype. We made deliberate cuts instead of half-built features, and focused everything on making one loop feel complete: capture → hold → resurface → act.