May 13, 2026 · BTCD Team
Building the USD Vault on IPOR Fusion
Bitcoin Dollar's USD Vault is live on IPOR. It's a leveraged carry trade that takes USDC deposits, deploys them into a delta-hedged sBTCD position collateralized on Morpho with a wBTC short, and produces dollar-denominated yield. The strategy mechanics are covered in our other posts.
This post is about the engineering surface beneath that strategy — what it took to make the vault shippable, what the IPOR Fusion framework provided, and what filled the remaining gap. Stated plainly: composable strategy infrastructure compressed what would otherwise have been a year of integration plumbing into a matter of weeks of design and configuration.
The Integration Surface
A vault running this strategy has to interact with, at minimum:
- Morpho Blue's isolated lending market for sBTCD/wBTC
- Morpho's flash loan facility for atomic rebalancing
- Uniswap V3 routing for fallback execution paths
- Aave V3 supply and borrow markets for the buffer carry
- The BTCD Portfolio's mint and burn interfaces for internally-priced trades
- The BTCDStaking ERC-4626 wrapper around sBTCD
- The Sky USDS and sUSDS contracts for the buffer sleeve
- Chainlink wBTC/BTC and BTC/USD price feeds
- Withdrawal queue accounting and request fee handling
- Role-based access control gating every privileged operation
- An execution surface that lets a monitoring service submit signed transactions without holding direct custody
Each of those is a multi-month engineering project on its own when implemented from scratch. None of them were written for this vault.
The Fuse Model
IPOR Fusion decomposes external integrations into composable adapters called fuses. A fuse is a small audited contract that knows how to perform exactly one operation against one protocol — supply USDC to Aave, borrow wBTC from Morpho, swap on Uniswap V3, take a Morpho flash loan, mint BTCD from USDC via the protocol's internal-RFQ minting interface. The vault holds a registry of approved fuses and executes strategy logic by composing calls into them. Building a new strategy reduces to: select the fuses, write the decision logic that determines which sequence to execute under which conditions.
A useful illustration is the dual execution surface for trades involving BTCD. The internal RFQ system — in which the BTCD Portfolio and the vault sit on opposite sides of the same trade and the protocol matches them directly to bypass slippage, fees, and MEV — has a corresponding pair of fuse families on the implementation side: a VaultMinting family for the internal path, and a Uniswap V3 family for the external fallback. Each rebalancing action — push capital into the loop, partial unwind, full unwind — has both a VaultMinting variant and a Uniswap V3 variant. The strategy logic chooses which to attempt; the fuses handle the encoding and execution.
The Diamond Architecture
Underneath the fuse layer, Fusion uses the diamond pattern (EIP-2535) — a contract architecture in which a single proxy address routes calls into many separate implementation contracts. The practical consequence is that the vault can grow new capabilities — additional fuses, new strategy modes, governance extensions — without forcing depositors to migrate to a new contract. The address users interact with never changes.
For a production strategy that needs to evolve in response to market conditions, integration availability, or new protocol features, this matters operationally. Adding the optional Aave-Morpho wBTC carry sleeve as a later iteration was a registry update, not a redeployment.
Permissionless Launch
Deployment on IPOR Fusion does not require whitelisting, governance approval, or a foundation's sign-off. Pick the fuses, write the strategy, deploy. That alone removed weeks of overhead that comparable frameworks would have introduced before any strategy work could begin.
What Was Provided, What We Built
| Component | Source |
|---|---|
| Protocol integrations (Morpho, Aave, Uniswap V3, USDS/sUSDS, BTCDStaking, VaultMinting) | IPOR Fusion fuses |
| Flash loan dispatch and atomic execution primitives | IPOR Fusion fuses |
| Contract upgradeability via the diamond pattern | IPOR Fusion |
| Deployment | Permissionless via Fusion |
| Custom sUSDS instant-withdrawal fuse | Co-developed with IPOR; upstreamed |
| Strategy logic, optimizer, monitor, safety overrides | This vault |
The pieces that are ours are the parts specific to running the strategy correctly: the monitor that polls on-chain state each cycle, the four-axis state classifier that maps the position into one of about two dozen named states, the deployment-fraction optimizer that decides how much capital to commit, the withdrawal-aware buffer floor that proactively deleverages when redemptions are queued, and the safety overrides that ensure no yield consideration can defeat an LTV or buffer constraint.
The pieces that aren't ours — and would have taken a year of integration work to be — are everything beneath.
The Custom sUSDS Instant-Withdrawal Fuse
One piece was missing from the existing fuse catalog when we started: a fuse for sUSDS that supported instant withdrawal. The standard sUSDS path did not permit synchronous in-transaction redemption — without that property, the buffer would have been operationally indistinguishable from a long-term position, defeating its purpose as the vault's exit valve.
We worked with the IPOR team to build it. The collaboration was focused and technically thorough — they were responsive throughout the design and review process, and careful about both the security implications and the broader interface design, not only for our use case but for what other strategies built on Fusion might need from the same primitive. Once the fuse was completed and integrated, IPOR published it in their official catalog as a standard building block. The team's responsiveness and rigor through the process is a meaningful part of why this vault shipped on the timeline it did.
A Modest Addition to the Catalog
The reason a buffer sleeve is valuable in a leveraged strategy is that it gives the vault a clean exit path: a yield-bearing place to retreat to when the carry spread is unattractive or when withdrawal demand is high. Every other strategy built on IPOR Fusion now has access to the same primitive for the same purpose. Vaults that previously had to hold idle, non-productive stablecoins to maintain liquidity now have a productive alternative they can fall back to without sacrificing instant access.
A modest contribution to the catalog measurably improves the structural safety of every strategy that adopts it. That's the right shape for ecosystem infrastructure: contributions compound, and the next team gets the benefit of the last one's work.
What This Enabled
The mechanics of the USD Vault — sBTCD as half-beta collateral, the leveraged market-neutral carry, the productive buffer, the autonomous rebalancer, the flash-loan-mediated atomic execution — are not, individually, novel. Each piece exists somewhere in DeFi already.
What's new is being able to compose them into a single, coherent, production-ready vault on a timeline measured in months, with a small team, while still giving careful attention to the safety properties that determine whether a vault should be trusted with real capital. That is the leverage the underlying framework provides.
IPOR Fusion gave us the substrate. We built the strategy. The vault is the result.
Bitcoin Dollar's USD Vault is live on IPOR today. The Morpho sBTCD/wBTC market is open and accepting wBTC supply. Phase 2 (cap scales to $10M alongside Yield Basis Hybrid Vault integration) follows in mid-May. Follow @BTCDOfficial for updates.
Vault: app.ipor.io/fusion/ethereum/0xf8f226da66244f89e70c5b5d1a5c5b0d505eb1d8