Pieter Wuille.

From Google to Bitcoin

Pieter Wuille is a Belgian computer scientist. He earned a PhD at the University of Leuven, where, his Blockstream biography says, he “focused on programming language design”, and went on to work for Google in Switzerland as a site reliability engineer, one of the people who keep its services running. On the side, an image format he wrote grew into FLIF, which he developed with Jon Sneyers and whose ideas later went into the JPEG XL standard.⁠^[1][2][3][4]

He first heard of Bitcoin at the end of 2010, “on a IRC channel about the Haskell programming language”. What drew him in was “this idea that you could have a currency defined by the internet, by nothing more than software.” His first real task came in February 2011, when Hal Finney, the cryptographer who had received Bitcoin’s first payment, set a challenge on BitcointalkBitcointalkThe Bitcoin forum Satoshi Nakamoto opened in November 2009. Much of Bitcoin’s early history, from the pizza offer to Satoshi’s last post, happened there., the Bitcoin forum. Finney posted a private keyPrivate keyThe secret number that lets you spend your bitcoin. Whoever knows it controls the coins, so it must never be shared., the secret that controls coins, holding 20 bitcoins: “Let’s see how long it takes for someone to input the key and collect the bitcoins.” The software had no way to import a key, so Wuille wrote one. “I don’t remember who won, but it wasn’t me,” he told CoinDesk.⁠^[5][6][7]

By late April 2011, a court later recorded from his evidence, he had joined Bitcoin’s maintainer team, the few developers allowed to merge changes into its code. For his first three years he worked on Bitcoin in his spare time.⁠^[5][8]

Rebuilding the engine

In February 2012 Wuille wrote BIP-32, a Bitcoin Improvement ProposalBIPA public document describing a proposed change to Bitcoin, or a standard that wallets can follow. for “deriving a tree of keypairs from a single seed”. A wallet that follows it can make as many addresses as it needs from one starting secret, and doesn’t need a fresh backup every time it makes one.⁠^[9]

His biggest early change was to how a node keeps track of money. Bitcoin’s first software recorded every transaction output ever created; Wuille’s rewrite, ultraprune, kept only the unspent ones, “an explicit UTXO setUTXO setThe unspent outputs a Bitcoin node keeps track of to check new payments against its accepted blockchain.”, in his words. It shipped in Bitcoin 0.8.0 in February 2013 with a faster database, LevelDB, and a judge later called it “a major performance improvement”. Within weeks it exposed a rule nobody knew existed. On 11 March a large block that 0.8 accepted was rejected by older versions because of a limit in their database, and the blockchainBlockchainBitcoin's shared record of every payment since 2009, stored in blocks that each link to the one before — which is what makes it a chain. split in two until big mining poolsMining poolA group of miners who share the work of finding blocks and split the rewards, so each gets paid steadily instead of rarely. went back to the old version. Wuille called it “an unintentional consensus failure”.⁠^{[10][11][8][12]}

In March 2013 he started libsecp256k1, a library for the maths behind Bitcoin’s digital signaturesDigital signature (ECDSA, Schnorr)Proof, made with a private key, that the owner approved a payment. Anyone can check it; practical forgery is infeasible under current assumptions., which prove that the owner of the coins approved a payment. A forum post by Hal Finney about a short cut that Bitcoin’s elliptic curveElliptic curve (secp256k1)The curve y² = x³ + 7, whose maths turns a private key into a public key that ordinary computers cannot reverse efficiently. allows had made him curious, so he wrote his own version of the maths “from scratch” instead of working inside OpenSSL, the general-purpose library Bitcoin used. It began as “a small hobby project” and came out “maybe a factor of 4 faster than OpenSSL”. Bitcoin Core began signing payments with it in February 2015 and, from version 0.12.0 a year later, checking them with it, “over five times faster” on typical PCs and servers.⁠^{[13][10][14][15]}

In September 2014 he found that OpenSSL could disagree with itself about whether a signature was valid, depending on the computer: “A specially-crafted transaction could have forked the blockchain.” He kept quiet, proposed BIP-66, a stricter rule for how signatures are written that closed the gap, and disclosed the flaw only in July 2015, after the rule took effect. In such cases, he said later, “moving too fast might draw suspicion, moving too slow might get exploitable.”⁠^[16][17][10]

The block size fight

In 2014 Wuille left Google to work on Bitcoin full time, as one of 11 co-founders of Blockstream, a company whose founders also included Adam Back, the inventor of hashcash, and Gregory Maxwell, another Bitcoin developer. From September 2014, the court noted, Bitcoin was his job.⁠^[18][1][8]

By 2015 Bitcoin’s developers were arguing over the 1 MB limit on the size of each block, which caps how many payments fit in it. Gavin Andresen, who had led the project after Satoshi Nakamoto, proposed starting at 8 MB and doubling every two years. That July Wuille proposed a slower rise, BIP-103, of 17.7% a year from 2017, roughly in step with improving technology. “Bitcoin’s advantage over other systems does not lie in scalability,” he wrote, but in “transparency, lack of need for trust in network peers, miners, and those who influence or control the system.”⁠^[19][20]

Andresen called it “much too conservative”, warning that the main blockchain could become “a settlement network, affordable only for large-value transactions”. Mike Hearn, a developer on Andresen’s side, said it was “virtually the same as keeping the 1mb cap”, and pointed out that it came from “a Blockstream employee”. Wuille stood his ground. A Bitcoin paid for by fees on large settlements would be “a much better outcome than what we have now”, he replied. “If Bitcoin needs to support a large scale, it already failed.”⁠^[21][22]

His own answer came that December at Scaling Bitcoin, a developers’ conference in Hong Kong. Segregated WitnessSegWitSegregated Witness: a 2017 upgrade that moved signatures into a separate part of each payment, making room for more payments., or SegWit, would move signatures, which “account for 60% of the blockchain data”, into a part of each transaction that older software never sees, and count them at a quarter of their size. He had long “dismissed this as a solution”, he admitted, until Luke Dashjr, another developer, found it could be done as a soft forkSoft forkA change that tightens Bitcoin's rules, so nodes that haven't upgraded still accept the new blocks., a tightening of the rules that older software still accepts, instead of a hard fork that older software rejects. “So I guess you can hear this coming,” he told the audience: “we can increase the block size with a soft-fork.”⁠^[23]

The same day Gregory Maxwell made SegWit the centre of a plan for Bitcoin’s growth, and Wuille was one of 58 people who signed a statement backing it. He wrote SegWit’s rules, BIP-141, with Eric Lombrozo and Johnson Lau, but it took until 24 August 2017, at block 481,824, for SegWit to take effect. “It’s been heavy,” he told CoinDesk that December. “I’ve been in the middle of it for longer than it’s been public.”⁠^{[24][25][26][27][28][7]}

Taproot and after

His next upgrade began over lunch. In January 2018 he ate at a diner in Los Altos, California, with Maxwell and Andrew Poelstra, Blockstream’s director of research. “While I briefly had to leave the table,” he recalled, the other two “had come up with a really cool idea”: TaprootTaprootA 2021 upgrade that added Schnorr signatures and addresses starting bc1p, making complex payments cheaper and more private., which lets a payment with complex conditions, such as several signers, look like an ordinary one when everyone agrees. Wuille co-wrote its rules, including BIP-340 for Schnorr signaturesSchnorr signatureA kind of digital signature Bitcoin added with Taproot in 2021. Several people can combine keys so their joint signature looks like one person's., which several people can combine into one, and Taproot took effect at block 709,632 on 14 November 2021.⁠^{[29][30][31][27]}

In 2020 he moved from Blockstream to Chaincode Labs, a Bitcoin development company in New York, saying he planned “to keep working on protocol improvements to Bitcoin”. On 7 July 2022 he asked for his key to be taken off Bitcoin Core’s list of trusted keys, ending more than eleven years on the maintainer team. He gave no reason; another developer, Ava Chow, noted that his last merge was more than a year old.⁠^{[32][1][33][8]}

He has kept writing code. BIP-324, which he co-wrote, encrypts the connections between Bitcoin nodes, and became Bitcoin Core’s default in April 2024. He also wrote key parts of “cluster mempoolMempoolThe waiting room for Bitcoin transactions that have been sent but aren't in a block yet.”, a redesign of how a node ranks unconfirmed payments by the fees miners can expect from them, which Suhas Daftuar, another developer, brought together for Bitcoin Core 31.0 in April 2026.⁠^{[34][35][36][37]}

He also answers questions. On Bitcoin Stack Exchange, a question-and-answer site, he has written nearly 2,000 answers and has the highest reputation of any user. His most popular answer works out how many bitcoins there will ever be: “If you sum all these numbers together, you get 20999999.9769 BTC.” In January 2025 he and Maxwell shared the Human Rights Foundation’s Finney Freedom Prize. The foundation called SegWit “the biggest Bitcoin protocol upgrade to date”, and said Wuille was “largely responsible” for it.⁠^{[38][39][40][1]}

Craig Wright and the courts

Craig Wright, an Australian computer scientist who claimed to be Satoshi Nakamoto, sued Wuille and other Bitcoin developers twice: in 2021 through his company Tulip Trading, which said they owed it a duty to write code returning bitcoins it said had been taken in a hack, and in 2022 with two of his companies, claiming copyright in the Bitcoin white paper. Wuille had doubted Wright since 2016, when Wright published what he presented as a message signed with one of Satoshi’s keys. Wuille found an old signature of Satoshi’s, already public on the blockchain, converted into a different format so that it didn’t look the same: “In short, the signature in the blog post proves nothing,” he said in a witness statement.⁠^[41][8]

In early 2024 the High Court in London tried whether Wright was Satoshi, in a case brought by an industry group, the Crypto Open Patent Alliance, and heard together with Wright’s claim against the developers. Wuille was the developers’ only witness. Mr Justice Mellor ruled that Wright was not Satoshi, and wrote that Wuille’s first witness statement had “turned out to be the most significant document in this Trial”: its account of when each feature entered Bitcoin was “used to devastating effect in cross-examination of Dr Wright”. Both of Wright’s claims against the developers ended in their favour that year, and the judge certified them as totally without merit.⁠^[8][42][43]