01
AethOS Case Study
- Role Lead Product Designer
- Tools Figma
- Type Institutional DeFi · RWA Routing
- Live site resplendent-sunshine-770cf4.netlify.app
The Concept
AethOS was conceived as an institutional-grade routing layer that connects idle treasury capital to real-world asset yield. Instead of static allocation across scattered protocols, an autonomous AI engine continuously scans on-chain liquidity pools and RWA issuers, US Treasuries, private credit, money market funds, to find the highest-efficiency path for every dollar routed. The design language reads like a trading terminal built for an AI co-pilot: a deep black canvas, confident data readouts, and a single core flow, select an asset class, simulate the route, execute.
The central design challenge was translating an inherently complex multi-chain settlement process, custody handoffs, MEV protection, multi-sig approvals, into something a treasury manager could trust at a glance. Every screen was built around live data: real-time route simulation, streaming yield, and a live execution ticker up top showing actual capital in motion, reinforcing that this is production infrastructure, not a mockup.
Micro-interactions.
Two fragments pulled straight from the live product, looping on pure CSS.
5,000,000 USDC
ERC-20
4,999,850 TBILL
+5.24% APY
Component-Driven Design
Every element starts from a single source of truth, a token-based system built for high-density financial data, live yield tickers, multi-sig signature flows, and route simulations that need to read instantly under institutional scrutiny.
A near-black trading canvas (zinc, not pure black) keeps dense route data legible. A single confident emerald drives every live, verified, yield, and interactive state, no secondary accent competes with it. Blue only shows up once, as a soft ambient glow behind the hero's 3D sphere, never on UI elements.
A geometric sans carries the institutional voice, a distinct display face gives headlines their edge, and a monospace family handles route hashes, latency figures, and SDK code so financial precision never gets ambiguous.
A 4-point scale (not strict 8-point, 12px and 28px both appear deliberately) keeps route cards, signature trays, and yield tables aligned across dense dashboards. Every transition shares one expo-out curve, giving AI-driven execution an instant, authoritative feel instead of a generic ease.
Badges
Cards
Panels
Pills