- Development
- Design
- Advisory
The First Marketplace for Industrial Chemicals
Commodity trading infrastructure
→View case study

Marinade launched in 2021 as the first liquid staking protocol on Solana. By the end of September 2026 it held 7.94 million staked SOL, and its web app had grown beyond staking into products like earning on USDC through Kamino. All of it lived on the web. Marinade wanted a mobile app for the people who already use it and for people who are new to it, with web3 assets and the ease of a web2 app. The app is a new product, not the web app on a smaller screen. It does not offer staking. Assets on Solana earn in it through Kamino. Marinade brought the core idea. We took on the mobile app from design through build to release. Product execution was shared between our team and theirs.
Designed and built the iOS and Android app from a blank page, as an extension of Marinade's team. Kotlin Multiplatform with native SwiftUI, plus part of the backend.
Registration creates a Privy wallet for the user. In the current build, every supported asset that is deposited moves on to Kamino and starts earning.
Every user gets their own Privy wallet and their own smart account. Funds move between the two and into Kamino. They do not pass through Marinade.
Proposed a new visual direction for the app on our own initiative and built most of its assets by hand.
Earning on Solana usually starts with a few decisions before any money moves. Which wallet, which protocol, which product. In the app those decisions are already made. Registration creates the wallet. A deposit of any of the five supported assets moves from the wallet to the user's own smart account and on to Kamino, where it starts earning. The user does none of that by hand. We designed these flows for the phone from a blank page, not from the web app's screens. Someone opening Marinade for the first time and someone who has used it for years get the same short path from sign-up to deposit to yield.


Midway through the build we proposed a new visual direction for the app, on our own initiative. The first drafts were ours. Every iteration after that was done together with Marinade, and it is the direction the app now has. AI is part of how we work on this project. For the visual identity we mostly set it aside. Icons, 3D assets and animations were largely built by hand in Blender, as one family that stays consistent on every screen.

The current build covers the first half of the product, getting in and earning. The second half is already in development, and it is about what a user can do with a balance once it is there. The plan is to let users draw on the value of their holdings without selling them, in three moments. Buying another asset in the app. Withdrawing to a wallet or a bank account. Paying, possibly with a card. In each one the option appears inside the flow the user is already in, at the point where it is useful, and the yield on what they hold is meant to offset what they use. Buying directly in the app with a card or Apple Pay comes first. We have already evaluated 8 on-ramp providers for it, and the stack is planned so that a user verifies their identity once across all of them.

When the app opens to the public, getting from sign-up to earning on Solana takes one short flow. Sign up, deposit, earn. The first build is approved for external testing on the App Store and Google Play, and the production build is now going through store review.
We’ve pre-loaded the context. Ask what we’re good at, and where we’re not.
Opens a new chat, context already written.
Bring a business problem, a protocol idea or a product that needs to ship. We'll help you find the right next step.
