Thomas Humphreys

This account is verified. Learn more

@thomphreys

funding coordination @ethereumfndn

Global Citizen thomphreys.com Joined October 2022

439 Following 437 Followers 431 Posts

Latest posts 5

Thomas Humphreys @thomphreys 2d

I spoke about how much harder it is for teams to find contract work compared to grants or FT jobs, and why that's a problem if we want builders to diversify their funding beyond grants

rfpsear.ch makes these opportunities easier to discover, helping funders and builders find each other

Great work by @karmahq_ bringing this to life!

Karma Verified account @karmahq_ 2d

x.com/i/article/2118…
View quoted post on X

Thomas Humphreys @thomphreys Oct 29

Can't imagine being the guy that held a trezor, got his address leaked so switched to ledger, then got spooked by a hardware hack, so deposited all on Binance

Spreek Verified account @spreekaway Oct 29

Unconfirmed, but possibly massive binance hack. Entire Binance wallet drained into new EOA. Contract call was funded by Tornado Cash

View quoted post on X

Thomas Humphreys @thomphreys Oct 25

Few thoughts as I wait to board my flight to Goa

I think crypto options are approaching an inflection point, with RFQ engines like @DeriveXYZ helping simplify access, but the bigger opportunity is embedding them into the dapps traders already use.

I see 3 things happening here:

1. Simplification drives adoption

Protocols like @DeriveXYZ and @ryskfinance are abstracting away the complexity of options, with e.g. Rysk replacing greeks and payoff curves with a simple UI showing your deposit, upfront APR, and expected outcomes (if/else statements).

2. Limit orders become a distribution channel

CowSwap did $3b in limit order vol past 3 months ($12b/yr annualized) yet users earned no yield on that flow. Imagine if users could sell puts or calls directly through the same UI, earning a premium for taking on an options obligation instead of simply placing a traditional limit order.

3. Curators and solvers manage the complexity

traditional limit orders can be cancelled at any time, whereas options lock you into an obligation until expiry unless you close the position earlier. Pricing that risk isn't trivial and most users won't know how to optimise for it.

This is where sophisticated solvers could compete to quote arbitrary strikes and qtns using hedged inventory, while curators help optimise for premium, risk, probability of the desired outcome, etc

The bigger bet is that limit orders could become a distribution channel to options while curators and solvers handle the complexity so users get better execution and optionality on their trades.

Anyway, I think this deserves a bigger write-up. Will publish a blog on this post-Devcon.

Thomas Humphreys @thomphreys Oct 6

this is underrated!

a core public good converts its adoption into revenue without restricting its free product - more of this, please

on a more personal note, would love to see privacy baked into this, esp for payroll + treasury ops

Safe Labs Verified account @SafeLabs_ Oct 6

Safe Pro is live: the most professional way to self-custody.

If self-custody has become critical for your team, @Safe Pro was built for you.

❇️ Oversee all your Safe accounts in one Workspace
❇️ Monitor transactions continuously with Security Hub
❇️ Implement policy controls on every transaction.

Your assets never leave your custody.

Check out our new website and get Safe Pro now: safe.global/pro

View quoted post on X

Thomas Humphreys @thomphreys Oct 1

The bigger point imo is that we need to validate these problems instead of making assumptions. If core devs really do have unlimited budgets and are moving too slowly, then lack of accountability is the problem.

I also don't think ZK necessarily makes client teams disappear. They may evolve into maintaining different guest programs / implementations, where preserving diversity there still matters. I generally like the move towards formally verified guest programs, esp considering that Lean has a much broader funding and development base outside Ethereum (tho Ethereum specs and implementations still need to be maintained by Ethereum itself).

I notice formal verification often gets framed as reducing one class of risk, but it also shifts risk deeper into the stack (spec, verification, proof sys assumptions). If more responsibility converges around one or two specs or implementations, that can create technical concentration, which can lead to governance / capture risk too. Someone still has to maintain, update and fund the canonical specs and guest programs, and whoever influences what becomes "canonical" ends up with a lot of the influence client teams have today.

Fede’s intern Verified account @fede_intern Sep 30

the EF, sooner rather than later, will have a clash with a relevant part of core devs. some of the new entities/foundations will probably support core devs.

ethereum has a clear governance problem that will be visualized thanks to three fronts: AI, post-quantum, and ZK. AI is making the size of the teams unnecessary, and it's removing funding from engineering projects.

the quantum threat is moving faster than ever, and the solution is also moving faster than ever thanks to AI. web2 projects can migrate fast to a post quantum solution but that's not easy in blockchains.

the third front, ZK, is gonna make most execution clients unnecessary. we will have one or two guest programs fully formally verified, and this will make most teams unnecessary.

the problem is that a big part of core devs have little to no interest in pushing for the new solutions. the slower the better. less risk, leave a bunch of engineers unmanaged with almost unlimited capital for a decade and this is what you get.

the EF, sooner rather than later, will have to push against those blocking what Ethereum needs. it's funny to me that a considerable part of CT believes the EF is still the problem to solve in Ethereum when it's the engineering core teams. the EF needs to be made even smaller and most of the engineering teams need to disappear.
View quoted post on X