FAQ: Team, Leadership & Structure

What exactly is Arthur's role — is he investing, advising, or lending his name? Who else is on the team, who runs engineering, and will the team be mining $FLOP itself?

Arthur Hayes is founder and CEO, involved day to day — explicitly not the passive investment or advisory pattern Maelstrom has applied across its portfolio. FLOP is his only active operating project.
He has no day-to-day responsibilities at Maelstrom; the fund is run by its head of investments, Akshat Vaidya. Maelstrom is his family office and it is wholly owned by him.
Funding: the investing entity in FLOP is Maelstrom, which Arthur owns outright, so the project is backed by him in substance even though the capital is committed through the family office rather than in his personal name. It is the sole equity backing, there was no outside round, and there is more than two years of runway.
Why: the centralisation of AI is the problem, and if a new silicon-based form of intelligence is being created it needs a currency and a coordination network — potentially the most valuable payments network ever built. Nobody is doing it right, and that is the gap.
Tokens: no pre-mine and no preference. Arthur's exposure sits inside the Labs per-block mint (8 $FLOP per block, 5.4% of year-10 supply), shared with the team, liquid on emission like every other block reward.
Team: Sergey Vidyuk, CTO (ex-Deutsche Bank, then technology architecture and infrastructure at BitMEX) leads all engineering; Shu Duan, Chief Strategy Officer (about 14 years in investment banking at Deutsche Bank, then crypto asset management at Aspen Digital; computer science degree) runs strategy and institutional and supply-side relationships. The team is small and runs hundreds of AI agents across development workstreams, which is what lets it ship what would previously have taken 20–30 people.
FLOP Labs will not run validators on the network. Whether FLOP Labs or Maelstrom will run miners of their own is not finalised; if they do, it is as ordinary participants under the published rules, disclosed openly, with no allocation and no side deal.
A technical diligence call directly with Sergey is available on request.
Headcount: the operating team is three. Arthur as CEO, Sergey Vidyuk as CTO and Shu Duan as Chief Strategy Officer. A fourth person is being hired to cover community and creator relations. Build work began around May 2026.
It is three people working full time, plus the agents they run — a deliberately small outfit, which is what makes building a performant chain with three people possible. Sergey worked for Arthur at BitMEX five or six years ago; the three have known each other in various capacities for years.

How much supply is soft-circled for launch? Who is engaged?

No signed commitments yet, and nothing described below is committed capacity.
Who is signing up: more than ten thousand inquiries from prospective miners since the project was announced, from individuals running a gaming PC at home to institutions running hundreds of enterprise GPUs out of data centres, with many GPU operators in between. The same rule applies to all of them — rewards follow compute actually delivered and verified, never idle capacity.
Application volume as at 15 September 2026, with the forms open since 18 August: roughly 9,000 miner applications, close to 4,000 validator applications and about 7,000 creator and community applications. The last precise triage, on 10 September, counted 8,357 miner applications of which 3,111 scored high quality, 3,425 validator of which 2,496 scored high quality, and 6,343 creator and community of which 3,117 scored high quality — 18,125 in total, 8,724 high quality.
These are applications, not units and not commitments. No stake has been posted and no hardware is reserved, and the quality score is internal triage rather than verification. Expressions of interest and GPU counts are different quantities and should not be read across.
Shape of the miner interest, counting high-quality applications only: about four in five plan five machines or fewer, and roughly three quarters name a consumer GPU with at least 16GB of VRAM. Enterprise cards appear in a meaningful minority, led by H100 and A100 with Blackwell-class parts behind them. Around 330 applications intend more than twenty machines. Validator interest skews to self-hosted over managed hosting by roughly two to one.
Indicated interest spans exchanges, market makers, prime brokers, crypto VCs, family offices, data-centre and GPU-cloud operators, validator-infrastructure firms and telcos, across mining, validating, broker and demand-partnership roles. Individual indications range from one or two rented units to clusters of 25–100 units, plus a mining-hardware group building Blackwell and Rubin capacity outside China.
Shape of the institutional supply now visible: one institutional participant has bought 160 RTX PRO 6000 Blackwell units outright to mine; most other institutional miners intend to rent, in a mix of H100, B200 and B300 capacity; several validator operators are preparing nodes, which need no GPU. None of it is contractually committed to the network and none of it is reserved — testnet is open and uncapped.
Target: about 1,000 H100-equivalents by end of testnet and about 1,200 at mainnet start. There is no testnet cap.
The binding constraint for enterprise-class miners is hardware access, not the protocol: data-centre GPU supply is tight, so an operator that owns hardware is better placed than one trying to rent. FLOP introduces miners to data-centre providers that still have capacity and publishes rental supply from the major providers at gpus.flop.finance.

Will you sign a mutual NDA, and does it restrict what Arthur says publicly?

FLOP will not sign a mutual NDA that restricts Arthur's ability to comment publicly.
Confidentiality obligations bind only what each company says through its own official channels. Anything an individual posts personally sits outside them, even if that person is an officer; nothing Arthur publishes personally can put FLOP in breach.
If a counterparty requires a restriction on Arthur, the alternative is to proceed without an NDA.
As a fully open-source project, FLOP does not sign NDAs with exchanges at all; it is happy to hold an informal listing or partnership discussion without one.
A narrow one-way confidentiality undertaking that protects only a venue's own commercial listing terms, binds the FLOP entity rather than any individual, and leaves Arthur free to comment on anything else is a different instrument from a mutual project NDA. Whether FLOP signs one is considered case by case and no general position has been taken.

Who controls the protocol after launch? Is the Foundation a Solana-style controlled foundation or a genuine Bitcoin-style handover — who sits on the board, is its reward stream locked, and what is the community's recourse if either entity is dissolved or compelled?

The intended model is the Bitcoin one: open-source code, a foundation that funds core development, and no permanent privileged proposer.
There are two Saint Vincent entities — FLOP Labs, the for-profit operating company, and the FLOP Foundation. Each mints 8 $FLOP per block as an additive subsidy alongside the 96 $FLOP block reward, halving on the same schedule and sunsetting after year ten. Together that is 1,955,232,000 $FLOP, 10.8% of year-10 supply, about 5.4% each. Both streams are earned block by block on the public schedule and are liquid on emission rather than vested or locked.
For the first halving (through block 63,072,000), only the Foundation can submit protocol-upgrade proposals — that is submit control, not vote control, and it sunsets at that boundary, after which anyone can submit. Approval is by on-chain referendum of token holders (Substrate OpenGov with conviction voting): a protocol upgrade needs 67% approval of votes cast, a support floor that starts at 50% of active issuance and falls to 15%, and a 14-day enactment timelock. Parameter changes within the approved runtime are rate-limited and cannot leave the safety envelope (at least 67% approval, 15% quorum, a 14-day timelock and the 1,200,000 $FLOP validator stake floor). Arthur has described the interim openly as a constitutional monarchy for the network's first two years: the Foundation proposes, the network decides.
Nobody controls the supply. The 4,400,000,000 $FLOP genesis pool and the per-second emission schedule are fixed in the open-source runtime at genesis; Labs and the Foundation cannot mint outside the 8 $FLOP per block each that the protocol pays them, and there is no treasury that decides allocations. The only way the schedule changes is a protocol-upgrade referendum clearing the thresholds above — which is the same position Bitcoin is in.
FLOP Labs will not run validators on the network, and it runs no agents on TechnoCore. Because changes are decided by referendum rather than by Labs, the testnet period is where anything that does not work gets fixed — after the genesis block there is no unilateral change by anyone, including the people who wrote the code.
The Foundation’s purpose is continued development on the Bitcoin Core model — using its block emissions to pay competent developers to maintain the network, which is the funding problem Bitcoin Core has never solved.
There is no investor allocation and no team allocation inside the genesis pool, and the per-block subsidy is minted like any other emission. The one genesis balance the Foundation does hold is the 800,000,000 $FLOP ecosystem reserve within the 4,400,000,000 airdrop pool — 20% released at genesis and 10% every six months thereafter — which funds creator and referral payouts, builder and security programmes and initial two-way liquidity; it is disclosed and is not a team or investor allocation.
One qualifier: at genesis the chain still carries a superuser key held by a multisig, and the one-way handover that removes it and routes upgrades through the governance track is specified but has not been executed. Until that handover completes, the accurate description is designed as a handover and not yet completed.
Board composition, entity governance and community recourse if either entity is dissolved or compelled are being documented in the governance section of the updated whitepaper.

How is FLOP Labs funded day to day, what is the runway, and does the protocol itself earn anything?

FLOP Labs and the FLOP Foundation each mint 8 $FLOP per block alongside the 96 that funds the network roles. It is an additive mint rather than a share of the 96, it halves on the same schedule, and it stops entirely after year ten. Together that is 10.8% of year-ten supply, 5.4% each.
It is liquid as it is emitted, exactly like any other block reward — no cliff, and no vesting schedule presented as a lock-up.
Until the chain is live there is no emission at all, so the operating company runs on funding from Maelstrom, Arthur Hayes’s family office, as it has since the start. There is more than two years of runway and Arthur has said he will cover any shortfall himself.
FLOP Labs builds and publishes the reference broker with its economics open so third parties can see the margin and compete; whether Maelstrom operates a broker itself is not finalised. FLOP Labs will not run validators; whether Labs or Maelstrom will operate miners of their own is not finalised, and if they do it is as ordinary participants under the published rules, disclosed openly.
The protocol itself earns nothing. The inference fee splits 85% to the miner and 15% to the validator and FLOP keeps none of it, so there is no protocol revenue line and no cash flow accruing to the token. The Labs and Foundation mint above is the only allocation that exists, and it is disclosed.

How will the Foundation's and Labs' tokens be held — would you use a third-party or exchange custodian for the treasury?

There is no pre-mined team or foundation treasury to custody at launch. FLOP Labs and the FLOP Foundation each receive 8 $FLOP per block as the chain runs — 10.8% of year-ten supply between them, 5.4% each — so their holdings accrue over ten years rather than existing as a balance on day one.
The one balance the Foundation does hold from genesis is the 800,000,000 $FLOP reserve inside the airdrop pool, released on a schedule: 20% at genesis and 10% every six months thereafter, fully released after four years. It is earmarked for creator and referral payouts, builder and security programmes and initial two-way liquidity at listing.
How those holdings are custodied — self-custody, a qualified custodian or an exchange custodian — has not been decided. It is being worked through alongside the listing conversations and no provider has been selected.
Qualified custodians have proposed a day-one integration so that institutional miners and validators can receive their allocations into custody through an onboarding link; that is under consideration alongside the reserve custody question and nothing is agreed.

Given the US–China tension over AI and chips, how is the business structured — which entities exist, where, and what does FLOP Labs actually do and not do?

FLOP Labs is a for-profit company that built a piece of open-source software and published it; the FLOP Foundation funds continued development. Both are Saint Vincent entities.
Neither entity sells tokens, provides compute, runs the validators that set the network's rules, sees the data that passes between agents and miners, or conducts inference. Labs and the Foundation are each paid 8 $FLOP per block by the protocol and hold no genesis allocation beyond the Foundation's disclosed 800,000,000 $FLOP ecosystem reserve.
Participation is a decision each miner, validator, agent and broker takes for itself: the rules are published, you follow them or you do not, and nobody receives tokens from FLOP — they are earned by doing work the protocol pays for. There are no special deals, side letters or private allocations for any participant, whatever its jurisdiction.
That is the whole answer to the geopolitical question: the network is not a Chinese, US or European project in any operational sense, in the same way that Bitcoin is not. Export-control and sanctions compliance for a particular GPU, data centre or participant is that participant's own determination, and FLOP does not issue eligibility statements for any country or person.