WhatTheHoot

• 

Inspiration

WhatTheHoot started with a simple idea: what if checking your vitals felt more like meeting a digital companion than looking at numbers on a screen? We wanted to combine computer vision, health-oriented sensing, and Web3 into an experience that is fun and immediately understandable. Instead of only showing heart rate and breathing rate, WhatTheHoot turns those measurements into an owl with a personality that matches your current vibe.

What it does

WhatTheHoot uses your phone's camera to measure your heart rate and breathing rate and then matches you with one of three owl companions: Chill Owl, Smiley Owl, or Hype Owl. After meeting your owl, you can take a selfie with your companion. The app automatically places your selected owl into the photo, and you can mint the finished image as a Solana NFT. The NFT is initially kept private in WhatTheHoot, and the user can explicitly choose whether to share it with the public community gallery.

How we built it

We built WhatTheHoot as an iOS application using Swift and SwiftUI. The app maintains the user's current measurements, companion, mood, energy, selfie, and NFT information in shared application state. The experience connects several technologies into one pipeline: Camera → Presage → Vital measurements → Owl matching → Selfie → Solana NFT → Snowflake-backed gallery The selfie system composites the selected transparent owl sprite directly onto the captured image before minting, so the companion becomes part of the actual NFT artwork rather than simply being an overlay in the UI.

Use of Solana

We use Solana Devnet to turn each WhatTheHoot experience into a collectible NFT. After taking a selfie, the app sends the composited image to our Next.js backend. The image and NFT metadata are uploaded to IPFS through Pinata, and the backend uses Metaplex tooling to mint the NFT on Solana and transfer it to the user's wallet. The NFT metadata records information about the user's companion, such as its pet type, mood, and energy. After minting, users can also open their NFT directly in Solana Explorer.

Use of Presage

Presage powers the core interactive experience of WhatTheHoot. We integrated the SmartSpectra SDK into our iOS application and use the front-facing camera to obtain heart-rate and breathing-rate measurements. Those measurements drive our companion-selection system. For example, a calmer combination can result in the Chill Owl, while elevated measurements can produce the Hype Owl, with the Smiley Owl representing the middle range. We intentionally present this as a playful companion-matching experience rather than a medical diagnosis or medical classification.

Reading breathing rate and pulse rate using Presage Swift SDK

Use of Snowflake

We use a Snowflake Postgres-backed gallery to connect the blockchain experience with a searchable application data layer. After an NFT is minted, we store information including its mint address, owner wallet, image URL, metadata URI, companion type, mood, energy, transaction signature, network, and sharing status. Privacy is part of the design: a newly saved NFT starts with is_shared = false. The user must explicitly press Share to WhatTheHoot Gallery before it becomes visible in the public gallery. The public gallery query returns only NFTs that users have chosen to share.

Snowflake Postgres DB Table nft_gallery

Data fetching off Snowflake Postgres

Snowflake Cortex AI implemented to analyze the breathing rate and heart rate, it will give a AI-generated suggestion when AI determines that either rate is abnormal.

Challenges we ran into

One of our biggest challenges was connecting several very different technologies into one reliable flow. We had to integrate real-time camera measurements in Swift, maintain state across multiple SwiftUI views, generate a composited selfie, upload assets to IPFS, mint on Solana, and then synchronize the resulting NFT with our database.

We also ran into practical issues around NFT metadata and image rendering, Solana Explorer links, multipart image uploads, database sharing state, and making sure an NFT remained private until the user explicitly chose to share it.

Snowflake no longer allows access to Cortex AI to its free tier users. So, we were not able to use Cortex AI, but used Snowflake to store database of nft into Snowflake Postgres.

After making new Snowflake account, we were able to implement Snowflake Cortex AI that analyzes the heart rate and breathing rate and give suggestion when detects abnormal high or low.

Accomplishments that we're proud of

We're proud that WhatTheHoot became more than a proof of concept. We built an end-to-end experience where a user can measure → match → photograph → mint → share from a single mobile application.

We're especially proud of combining Presage's camera-based measurements with personalized owl characters, creating the companion selfie dynamically, minting that image as a real Solana NFT, and building a privacy-aware gallery where users control whether their creation becomes public.

What we learned

We learned a lot about integrating technologies that normally live in very different ecosystems. On the mobile side, we worked with SwiftUI, camera capture, image compositing, and real-time measurement data. On the backend, we worked with Next.js APIs, IPFS, Solana and Metaplex, and database persistence.

More importantly, we learned that emerging technologies become much more approachable when they're hidden behind a simple experience. A user doesn't need to think about IPFS hashes, blockchain transactions, or database records—they can simply meet their owl, take a picture, and decide whether they want to share it.

What's next for WhatTheHoot

Next, we'd like to expand WhatTheHoot beyond the current three owl personalities with more companions, expressions, accessories, and collectible traits. We could also make companion matching more sophisticated by incorporating additional signals available through Presage and tracking how a user's companion changes across different sessions.

On the Web3 side, we'd like to expand the gallery into a richer social experience where users can discover community Hoots and build collections of their companions. Ultimately, we see WhatTheHoot as a playful bridge between computer vision, digital companions, personal data experiences, and on-chain collectibles.

Built With

• godaddy
• next.js
• presage
• snowflake
• solana
• swift
• typescript