desnake

50 posts

desnake banner
desnake

desnake

@desnakeee

pm @gearboxprotocol

Katılım Eylül 2023
385 Takip Edilen127 Takipçiler
desnake
desnake@desnakeee·
Great reads by @0xPrince & @SilvioBusonero on how DeFi market structure shapes lending 1. Lenders expect instant liquidity 2. Borrowers don’t value duration 3. Duration doesn’t clear Fixed rates need new users: lenders who will warehouse duration and borrowers who demand it
Prince@0xPrince

x.com/i/article/2000…

English
0
0
4
131
desnake
desnake@desnakeee·
@0xmikko_eth e.g. MEV, Stakehouse, Avantgarde, AlphaPing, and Gauntlet all have significant srUSD/USDC caps for a single (oracle, LTV, liquidation premium). Vault structure incentivizes concentration of allocations and can nudge curators toward riskier decisions.
English
0
0
1
33
desnake
desnake@desnakeee·
@0xmikko_eth One thing that’s always interested me: how come there’s only one risk parameters config for most collateral/debt pairs on a Vault-based lendings?
English
1
0
2
273
0xmikko.eth
0xmikko.eth@0xmikko_eth·
Lendings are not vaults—stop the deceptive mimicry I cant agree with Paul: Lendings are Web3's banks, permissionless protocols like Aave or Gearbox, where deposits earn yield from collateralized loans. Vaults are speculative funds that automatically optimize across various risk strategies, masking volatility behind "yield" facades. During the xUSD depeg chaos, Morpho's vaults stung depositors with meager yields amid the same dangers, turning deposits into toxic traps. This flawed setup increases risks: borrowers can exit positions quickly to protect their margins, but lenders stuck at 100% utilization are stranded, unable to withdraw, and left holding the bag. Paul's perspective reveals the truth: vaults are not safe retreats. They are high-stakes gambles posing as deposits, undermining confidence by copying bank-like assurances without real safeguards. Education is only a temporary solution for poor design. Real growth requires honesty: vaults should acknowledge their fund-like risk, rather than pretending to be safe harbors and mimicking lending. Protocols that mix them create Frankenstein finance: risky, opaque, and exclusive to elites.
Paul Frambot 🦋@PaulFrambot

A vault is comparable to an onchain fund, and just like traditional funds, some will perform well and others won't, but this is what we must accept and mitigate if we want to build a truly open and decentralized system. The fact that only 1 out of ~320 vaults on the Morpho App had limited exposure to xUSD isn't evidence that the model doesn't work — quite the opposite. Morpho's isolated market + vault model meant all 319+ other vaults and their depositors, each with different risk profiles, had zero exposure. Many assume losses equal system failure. However, in open financial systems, losses are a natural consequence of risk-taking, even when systems operate exactly as designed. As an industry, we should not fall for the fallacy that yield is risk-free. Instead, we should focus on better surfacing and educating about risks — ourselves included. For DeFi to be the backend of finance and scale to trillions in lending volume, lending infrastructure must remain separate from risk management. We firmly believe that this open and permissionless approach is the right one, and how DeFi was intended to work.

English
12
4
68
12.6K
desnake
desnake@desnakeee·
@PTomsky @invariantgroup @solettyy @StreamDefi Wen hcl-curated pools on Gearbox to show how it’s done? Jokes aside if you have thoughts on how to improve the system, not just for Invariant but for curated lending in general, I’d love to chat about possible ways to collaborate.
English
1
0
1
168
Phil.xyz
Phil.xyz@PTomsky·
@invariantgroup @solettyy @StreamDefi Look, i agree with you in general. But it’s not FUD. It does not matter, how liquidity is pulled - directly or via “quota” you considered its normal to have 90% exposure to Stream. Would love to talk to you outside the “noise”. My dms are open, can accommodate.
English
2
0
3
1K
Invariant Group
Invariant Group@invariantgroup·
We'll be very clear regarding our position when it comes to @StreamDefi and other considered in high risk of defaulting, we do not care about CT fud, we do not care about the noise, we build conviction based in proper DD and of course we can make mistakes in configuring IRM's, some collateral specific rates, etc... but thankfully @GearboxProtocol provides the proper tools for us to readjust and adapt to market conditions, our advice for everyone and other so considered "risk curators", if your actions are determined by noise from random people, you are not a risk curator, you are an opportunistic.
English
2
1
22
11.5K
desnake
desnake@desnakeee·
If liquidation at market price would cause bad debt, the protocol checks again using the fundamental price. If the position would be healthy under that valuation → 🚫 liquidation doesn’t go through Basically, a safety net when markets go nuts.
English
1
0
2
425
desnake
desnake@desnakeee·
🧵 DeFi lending is built on oracles. Get them wrong and you'll get hacks, bad debt, or drained pools. 💀 Here’s how most DeFi oracles fail and how Gearbox built a system that fixes the problem. ⚙️
English
2
4
20
2.1K
desnake
desnake@desnakeee·
@definikola That works until mHYPER doesn't have leveraged mHYPER as its backing. Looks like for now it doesnt
English
0
0
1
44
desnake
desnake@desnakeee·
@definikola 3. Holding mHYPER itself is ok, supplying USDC to credit borrowing against mHYPER is even safer => no problems
English
1
0
1
26
definikola
definikola@definikola·
BA Labs has no exposure to: - mHYPER - yUSD - rUSD mainly due to recursive collateral risk (collateralization by their own tokens)
CBB@Cbb0fe

Hey @MidasRWA and @hyperithm You are promoting mHYPER as a "liquid yield token". I would be very curious to see what happens if everyone tries to withdraw from this liquid yield token. Thank you.

English
3
3
23
5.8K