Timechain
Search Donate
Sponsors
Cake Wallet, private Bitcoin and Monero walletSideShift.ai, no-signup crypto exchangeOrangeRock, non-custodial bitcoin spot and perps tradingBraiins Hashpower, buy and sell Bitcoin hashrateLayer Two Labs, drivechains for Bitcoin
S17 · E14 · 1:42:23

S17 E14: Kristian Csepcsar on the History of Bitcoin Mining, Braiins Hashpower & the Centralization Nobody Talks About

20 Mar 2026 · by Vlad Costea · 28 min read
Published at block 941,368
MP3 (191 MB)

The audio keeps playing while you browse the site.

Read the transcript17,885 words, every line links to the minute it was said

Ask this episode

Embed this episode

Paste this on your site or blog. It is a small player that streams from here: no trackers, no cookies, no ads.

Preview the player

Editor’s note: this conversation was recorded in person at the undisclosed Braiins office in Prague, most likely in March 2026, when the Braiins Hashpower marketplace was about to complete its first month of public beta. Figures quoted (hash price, Bitcoin Cash profitability, market shares) reflect that moment in time.

There are podcast episodes you record remotely, and then there are episodes where the sponsor flies you into an undisclosed European location, hands you a piece of whale salami, and lets you loose in their studio for an entire day. This is the second kind. Braiins Hashpower had been sponsoring the show for five or six episodes, and part of the deal was that I would come to Prague and record a whole series in person: Kristian Csepcsar first, then Summer Meng of Bitmars, then the Firefish people, then Lukas (whose tweet about OnlyFans I promised to get on record), then Grafton from Vexl. All of it in the same studio, all of it with the same slightly chaotic energy.

Kristian is the head of marketing at Braiins, a man I introduced as the Chief Meme Officer, which he immediately loved. He travels the world with a drone, a film camera, and a giant inflatable Bitcoin (which I described as a huge cheese roll, a reading he also accepted), documenting mining operations and collecting use cases. He had just returned from beyond the Arctic Circle in Norway, where a company called GreenTech runs a two megawatt mining farm that supplies heat to a giant office building at minus twenty Celsius.

And because he is the kind of person who brings souvenirs, he handed me salami made out of whale. Norway is one of the last countries where you can eat whale; restaurants serve whale steaks, and whatever is left goes into the salami. I asked the obvious question. He said he could not confirm which parts went in, but that there was, statistically, a very high chance. Reader, I ate it anyway.

The book, the audiobook, and the toner argument

Kristian is the author of The Ultimate Bitcoin Mining Guide, a genuinely beautiful object with a foreword by Summer Meng (yes, the ASIC Queen, and yes, we get to that story later). All Braiins books, nine of them at this point, covering everything from mining economics to coding Bitcoin in Rust, are free to download from Braiins Books in exchange for a newsletter subscription you may later cancel, which Kristian calls a fair price. There are audiobook versions too, and this one is narrated by none other than Guy Swann of Bitcoin Audible, because you have to do these things properly.

The book is written in an unusual format: ten sentences per page, designed to read like a Twitter thread, finishable in about three hours. It also contains memes, which created the best sales argument for the print edition I have ever heard. The paperback costs around ten dollars on Amazon with next-day Prime shipping, and if you tried to print the free PDF yourself, you would spend more than that destroying your toner cartridge on the memes.

It is a nonprofit project for Braiins; the goal is volume and education, not margin. Also worth noting: only two people in history have pronounced “Csepcsar” correctly on the first attempt. I am one of them. The other eight billion of you should reconsider your lives.

Sixteen years of Braiins, or the oldest Bitcoin company in existence

The overarching theme of the episode was a short history of Bitcoin mining, and there is no better guide, because Braiins’ own story is most of that history. Most people know the company through the Slush Pool era, since that is when the absolute number of small pleb miners peaked around 2017 and 2018. But the naming story is better than the brand. The pool started in late 2010 as “Bitcoin.cz mining pool”, the .cz being the Czech ending.

Its creator signed his emails with his nickname, Slush, so people started calling it “Slush’s pool”, then Slush Pool, and the name stuck informally for years. In 2013 Slush wanted to focus on building Trezor and could not scale the pool alone, so he invited two friends to take over. Those two friends are the two I’s in Braiins, a name Kristian says he has to work with every day and that makes his life very difficult. The company officially renamed the pool to Slush Pool only in 2016, right before the bull market that cemented it in everyone’s memory.

The result of all this is a remarkable claim: Braiins may be the oldest Bitcoin company ever, not just in mining. There was no concept of a Bitcoin company before Slush started his pool in 2010. The pool has mined 1.3 million BTC over its existence. And as I noted on air, the whale salami suddenly made sense as a flex.

I tried a “there is no I in Braiins” joke. It failed, because there are literally two I’s in Braiins. Kristian, gracious as always, confirmed the company contains many very smart and very opinionated people, so there is no shortage of I’s.

Hashpower, the lottery, and the best marketing that money cannot buy

Since Braiins Hashpower sponsored this season, we spent a while on it, and the timing was absurdly good. The service lets you rent SHA-256 hash rate on demand and point it wherever you want, including at solo mining endpoints for a shot at a full block reward. At the time of recording it was two days away from completing its first month of public beta, and in that first month someone had already used rented hash rate to hit a solo block through Braiins Solo. The market went crazy: they had to deploy an additional exahash of capacity, were selling out daily, and were suffering the kind of stability issues that only sudden success creates.

Kristian’s framing of solo mining as a lottery is more honest than most gambling marketing you will ever see. His pitch, and he has about thirty orders of his own running, is that the odds beat the Powerball by orders of magnitude.

Mine with one petahash for a day and your odds are roughly one in seven thousand, with a new roll of the dice every ten minutes instead of one drawing per week, and a live view of how close your shares get to the difficulty target. It is continuous, it is engaging, and your money goes to securing Bitcoin rather than to a gambling company, so you get to feel virtuous while degenerating. Mine responsibly, as they say.

This is where I learned, with some dismay, that at the time of recording the hash price on Bitcoin Cash was slightly above Bitcoin’s, something like $46 against $44, and that solo gamblers were renting hash rate to mine BCH because the odds were better. The global BCH hash rate was around five or six exahashes against Bitcoin’s one zettahash, so on the marketplace you could buy a fifth of the entire Bitcoin Cash network’s power for half an hour and take a serious swing at a block.

I will repeat here what I said in the studio: call me whatever you want, but in a world where Bitcoin is winning, it should not be more profitable to solo mine Bitcoin Cash. It was, at least briefly. Hash price was at all-time lows and the hash rate still was not turning off, a combination that tells you something uncomfortable about who is mining and why.

And it is not only gambling. Kristian gave a clean B2B example: hosting companies with contractual uptime obligations, say 90%, can cover an unscheduled maintenance window by renting hash rate for twelve hours instead of paying fines to their clients. Four days before we recorded, someone had also rented three petahashes and pointed them at Public Pool, which then found a block.

The solo mining experiment was producing winners across venues, and as Kristian admitted, that first block did more marketing work than anything they could have built in the following months. For solo mining especially, provable track record matters enormously: nobody wants to point hash rate at an endpoint that has never produced a block, because you cannot tell the difference between bad luck and broken infrastructure.

The AI pivot and the 2017 déjà vu

One of the more sobering segments came from Kristian’s chief of commercial, who had just returned from a conference with a clear message: nine out of ten big publicly traded miners are pivoting to AI. Kristian’s bet is that most of them will be out of mining completely by the end of the year, that the AI boom continues for another two or three years, and that some kind of bubble eventually bursts. The perverse part is the market mechanics.

A public miner does not even need the AI business to be profitable; announcing the pivot alone pumps the share price. Meanwhile a miner that posts good numbers and stacks sats gets nothing. If you want to raise capital to build, you have to be on the AI wave, which is exactly how 2017 felt, when companies added “blockchain” to their names and watched their stocks levitate. History does not repeat, but it does rebrands.

Against that backdrop, Braiins launching a consumer hash rate marketplace is almost contrarian. The comparison Kristian drew: a BitAxe gives you one terahash for its whole lifetime. For the price of a BitAxe (the cheapest one with a terrible fan runs about $400) plus a month of European or American electricity, you can rent orders of magnitude more hash rate on the marketplace, with adjustable budget, duration, and speed. One petahash for two hours cost about six dollars at the time, and one petahash is a thousand BitAxes.

For completeness, since I asked: BSV’s entire global hash rate is so small that you could 51% attack it for a few dollars, and Braiins alone controls roughly four times it. Brutal, and also a decent live demonstration of why minority SHA-256 chains are a security punchline.

The actual short history of Bitcoin mining

Now to the history proper, which Kristian tells well because he wrote a whole chapter on it. His view is that the mechanics of mining never change (it is still rolling dice against SHA-256), but who mines, at what scale, and with what tools changes constantly, and you cannot understand where mining is going without the archaeology.

Satoshi’s original design assumption, one CPU one vote, was naive in the most endearing way. There is a real email from Satoshi to Laszlo Hanyecz, the first GPU miner, essentially asking him to please stop, because GPUs were shutting regular users out of the coin distribution. Satoshi added that he was not a socialist and would not tell Laszlo what to do, but the plea was there. Months later, the very dynamic Satoshi was worried about accelerated.

The Prague part of the story deserves retelling too. Slush learned about Bitcoin through an internal email at the open source peer-to-peer file storage project he worked on with Zooko (yes, that Zooko), who one day in 2009 or 2010 told the whole team to check out this cool thing called Bitcoin. Slush started mining, then had the pivotal thought: instead of competing against everyone for rewards, why not join forces with friends and split proceeds by contribution? That is the invention of pooled mining, announced on Bitcointalk in late 2010. The pool mined its first block on December 16, 2010, after about a month of trying.

One correction Kristian insists on, for the curious minds: Slush’s was not chronologically the first pool. About a week earlier, someone launched a pool called Pudding Pop (spelling uncertain, pronunciation impossible), but it had no public stats, worked badly, nobody connected to it, and its website died right around the time Slush mined the first block. Braiins therefore claims the title of first successful implementation that stood the test of time, which is fair. History simplifies, and nobody remembers a website that did not work. Someone should launch a Pudding Pop meme coin.

The hardware progression went as you know: the GPU boom of 2010 killed CPUs (pools were partly invented to keep CPU miners in the game a little longer), FPGAs came and went, and in early 2013 the ASICs arrived and everything went absolutely crazy. The first ASICs came from Canaan, whose CEO is an original gangster coder and Bitcoin lover, and who beat Bitmain to market by almost a year. Canaan then proceeded to be so bad at business that, thirteen years later, publicly traded on Nasdaq and perpetually near delisting, they hold maybe two percent of the global market. Bitmain arrived at the end of 2013 and has held around 80% of the hardware market ever since.

And then there is the Butterfly Labs tragedy, which I have heard from multiple survivors independently. GPU miners who had stacked serious bitcoin did the napkin math on the first announced ASICs (a hundred times my hash rate means a hundred times my bitcoin), prepaid for machines in BTC, and then watched the companies deliver catastrophically late or not at all, while the ASICs that did ship spiked the difficulty so hard that even delivered machines never recovered their purchase price. A coordinated wealth transfer executed by people who had no idea the others existed. The first halving landed in the same era for good measure.

The accidental 2% fee

My favorite piece of pool lore from this episode is the origin of the industry-standard fee. Slush originally wanted to launch the pool with no fee at all; he did not expect the model to survive and just wanted to keep CPU miners in the game a while longer. He got backlash on Bitcointalk: a free pool would centralize power. So he put a fee on it, and, this is the part that is half urban legend and half confirmed by the man himself, he picked two percent essentially at random. It felt about right.

Two percent became the ceiling of the entire industry’s fee structure, sliding down with size. Then the Chinese pools entered with FPPS, which pays miners far more predictably, and started the price war the Chinese industry always starts. From an already razor-thin margin, everyone had to go lower, and Slush spent years regretting that he had not started at five.

FPPS economics are brutal: the pool must finance stable payouts from reserves while collecting an effective fee (after big-miner discounts) of maybe half the public number. Sixteen years later, the whole industry still lives with baggage created on a whim by people who thought the experiment would be over in a few years. The pool’s birthday is at the end of November, by the way. I demanded a meaningful party, not a Tuesday beer. Kristian promised to seriously consider it.

The Linode hack, and how catastrophe built Trezor

In 2011 the pool was hacked for about 3,000 BTC through Linode, the third-party hosting provider (possibly an inside job; the vulnerability was theirs, not Slush’s configuration). In today’s numbers the sum is insane. What Slush did next defines the era: he paid everyone out of his own pocket, deciding the alternative was the end of the business. The incident directly set three things in motion.

Slush decided bitcoin needed a safer way to sign transactions, which became Trezor. He decided the pool was too much to run alone, which brought Braiins in. And the new team made the architectural decision that nothing would ever live on third-party servers again: they built their own hardware, encrypted it in-house, in the office, and shipped it to data centers around the world.

If you listen to old interviews with Slush from that period, the stress is palpable: alerts on his phone at all hours, leaving dates and the opera to run home and fix the pool under attack. The Braiins founders inherited the same lifestyle, laptops everywhere, permanently on standby. Wild West times, well before Kristian joined in 2018 from a KPMG IT audit job, which he left to make memes about terrible hash price. His favorite is a medical pain chart where the hash price entry is a man lying dead underground, drawn when the hash price was 0.06, which at recording time would be an improvement.

XT, Taproot, and the most political pool

Braiins is maybe the most political mining pool in Bitcoin’s history, and I say that as a compliment. In 2015, when Gavin Andresen published Bitcoin XT with BIP 101’s block size increase, Slush Pool was the first pool to signal support, and got DDoS attacked for it. In 2021, the first block ever mined with a Taproot support message was mined by Braiins. The pattern is a company of bitcoiners first, who happen to work in mining because they are good at software, versus much of the industry (and Kristian says this without malice) where leadership is business first and Bitcoin is a revenue stream.

That culture runs into the present: Lukas, whose literal job title is Braiins Superman, was at the time doing a deep analysis of BIP 110 (formerly BIP 444, a numbering change whose politics we could not untangle on air), planning a podcast series interviewing both sides. Kristian’s position on the whole affair was the most reasonable thing said about it anywhere: the truth is found in discussion, and what he sees online is not discussion but a fight, which convinces nobody. Pools, in his view, have an educational duty toward their miners, who are mostly businesspeople with opinions on Bitcoin but no capacity to read code, and then the miners vote with their hash rate if they disagree. As simple as that.

For the record, Lukas also likes BIP 300, which is controversial in some circles and very good for miners in all of them.

Where the hash rate lives, and why Russia and China stayed out of reach

A common misconception worth killing: mining pools do not own the ASICs. Foundry and Marathon run their own operations alongside their pools, but Braiins’ model is the original one, captured in the pool’s first slogan, “mine as one”: clients own the machines, the pool combines them virtually, rewards split by contribution. Where the clients are, though, is its own story. Do not quote him precisely, but Kristian put about 80% of Braiins’ hash rate in North America.

Russia and China were never penetrated, for reasons that say a lot about both. In China, being Chinese with a Chinese entity is table stakes, and merely getting a domain and a server placed in-country is an exercise in insanity. In Russia, the experience (he stressed it was theirs, not a universal claim) was that nine out of ten conversations ended with the farm operator asking for a secret personal kickback on the firmware dev fee in exchange for installing Braiins OS on machines owned by absentee investors.

Braiins never did it, because it works with publicly traded companies and its reputation is the only thing keeping it in the game long term. Europe barely mines. Ethiopia and Paraguay run the same playbook: entice you with cheap rates, let you build megawatts, then hike taxes, with unscheduled twenty-minute blackouts in city centers as a bonus that wrecks the uptime your profitability model depends on. Meanwhile Texas holds roughly a third of global hash rate and is, politically, the best protected place on earth to mine. Come and take it.

Braiins was also about to launch its own six megawatt site in Houston, running its own stack, for a delightfully simple reason: saying your firmware mines more bitcoin is one thing, publicly showing it on six megawatts is another.

The pitch: no KYC, Lightning payouts, and firmware with a brain

Since my short mining career ran through Braiins and Luxor back when KYC was not a thing, I asked for the home miner pitch. The answer: there is no KYC anywhere in the stack. Pool, solo, and Hashpower all need just an email address; Hashpower additionally uses Telegram for two-factor on orders, an admittedly vibe-coded public beta solution to be replaced later.

Compare NiceHash, which wants your passport. Kristian’s reasoning is the part I wish more companies internalized: Braiins already custodies some miner funds and fends off attacks daily, so why would it voluntarily add a database of the names and addresses of OG bitcoiners who mined tens of thousands of coins in 2011? Not wanting the honeypot is the security policy.

The other killer feature for small miners is Lightning payouts, which Braiins pioneered among pools about two years earlier. Thousands of users run one to five terahashes (a BitAxe or two), mine for the freshly minted sats and the beer budget rather than the lambo, and withdraw over Lightning with far better opsec than on-chain dust.

Then there is the firmware story, which I think is the cleanest demonstration of what Braiins actually is as an engineering company. Walking through their hallway, past the Antminer S19 XP they use for testing, Kristian explained that stock firmware handles overheating by shutting the machine down (or, in the newest generation, hard downclocking). Since profitability models are built on two variables, energy price and uptime, a machine that dies at temperature is a bottom-line disaster.

Braiins OS instead scales power consumption down in steps, say 100 watts at a time on a 3 kW machine, until the chip temperature stabilizes, then scales back up when conditions improve, whether because night fell or because someone fixed the intake airflow. My reaction on air was the correct one: why are you presenting this as a feature? This sounds like the first thing any firmware would ship with. It is not. Bitmain, with 70 to 80% of the hardware market since 2014, has never built it.

Bitmain, and the graveyard of challengers

The Bitmain segment is where Kristian’s diplomacy thins. Their chips are excellent and their innovation speed real, but their corporate attitude is legendary: whenever a competitor gets close, they slash prices and release a new generation, let the challenger bleed, then raise prices again. The anecdote of the episode: when Bitmain announced Summer Meng’s Bitmars as a hash rate ambassador (an exclusive tier, he claims, that costs $40 million), the top replies were from VoskCoin and Red Panda Mining, the two biggest mining YouTubers alive, publicly complaining about broken machines and refused repairs.

There are maybe four mining YouTubers on the planet; keeping them happy costs one free machine. Kristian says he has heard from sources that Bitmain’s internal view is that YouTubers should pay them for the collaboration. They know you will come back and buy anyway.

The challenger graveyard proves the point. Intel tried chips. Samsung invested in Auradine, the American ASIC maker, which announced a full pivot to AI the week before we recorded. Block has been working on a full miner for years, and Kristian, who insists he is a fan and speaking from his own hard experience, would bet they either do not release this year or send a small test batch to Core Scientific and call it a launch.

Hardware is hard; Bitcoin mining hardware is harder; and the chicken-and-egg problem is vicious, because nobody wants to be the first testing mouse for an unproven manufacturer, and without customers you cannot build the repair and support infrastructure that miners require. His proposed unlock is a merchant chip supplier, since the chip is the hard part and system design around it is comparatively easy (heaters, boilers, all the heat reuse use cases). The Chinese manufacturers will never do it, because they all want to be Apple: the margin is on the whole machine, ideally one that breaks so you buy another.

The most interesting contender is Bitdeer, Jihan Wu’s company after the two Bitmain founders split a few years back. Bitdeer is one of the biggest self-miners in the world, started building machines for its own farms, and only recently began selling publicly, with the usual late deliveries and cancelled orders, and with heavy Tether investment behind it. Kristian fingers-crossed for them, partly for the irony: the challenger who finally dents Bitmain’s dominance would be Bitmain itself.

A word on Jack, screens, and how not to take feedback

I will summarize my Jack Dorsey segment briefly, because I promised Jack (who is warmly invited on the show) that I would not roast him too hard. The pattern: LDK, the Rust Lightning library by Matt Corallo that he funded, failed to take share from LND, Eclair, or Core Lightning. Bitkey, the hardware wallet, shipped with no screen, in a market where Ledger’s first screenless device failed and was saved by the Nano S (S stands for screen) and BitBox’s first screenless device failed and was saved by the BitBox02. Ten extra dollars of bill of materials, ditched by Silicon Valley disruption logic.

Kristian’s related story is the cautionary tale. He posted detailed feedback on Block’s Proto miner from the perspective of a typical nontechnical miner (why are the powerful fans only on one side, where is the redundancy, maybe explain this on the website), and two high-level Block employees responded by publicly mocking him as a disgruntled firmware competitor threatened by their product. His point stands regardless of the product: either do not engage, or engage with a thank-you and fix the website. Going into a critic’s mentions as a publicly traded company is the worst option, and it guarantees the story gets told in a podcast article years later.

I offered my Coldcard parallel: marketing a device as open source while the firmware license forbids commercial reuse is not open source, it is source available, and the license page itself says so. Their response was to block me and diagnose me as a Trezor shill. When the idea is uncomfortable, attacking the person is always easier. It is also always an admission.

The centralization conversation nobody is having

This was Kristian’s rant, and it is the most valuable section of the episode, so I will give it the space it deserves. Everyone talks about mining pool centralization, because pools are old, understood, and easy to yell about, and because a loud new pool entered the market whose entire positioning is “all pools bad, we good”, which is a marketing strategy as much as an architecture. Kristian’s argument is that the stack has worse single points of failure that get zero airtime.

First, firmware. Bitmain runs 80% of the hardware market and ships closed-source firmware. His explicitly hypothetical thought experiment: a manufacturer controlling most of the world’s machines, which also runs a competing pool, could silently discard a golden nonce found for another pool, and nobody would ever see it happen. He is not accusing anyone of doing this.

He is saying the architecture permits it, that firmware is the critical layer of the mining stack, and that it should be open source in some capacity. This is the same Bitmain, he notes, that claimed the S9 could not physically run overt AsicBoost, until Braiins published proof that it could, at which point the capability appeared overnight, with 13% energy savings. Where that capability had been pointed before is a question that answers itself.

Second, management software. Foreman owns something like 80 to 90% of that market, sits between your firmware and your pool, knows everything about your farm: your energy programs, your exact revenue, your structure. Kristian knows and likes the team and has partied with the founder; there is no conspiracy here, and that is precisely the point.

A monopoly that dominant is an intelligence goldmine even if its owners are angels, because the data exists to be subpoenaed, breached, or infiltrated. And the switching costs are brutal in a way pool switching is not. You can reconfigure ten thousand machines to a new pool in two minutes.

Management software is baked into your accounting, your reporting, your site maps, your trained staff; switching takes weeks to months on a big farm. Pools have been debated to death for a decade; he has never once heard a public conversation about the management layer. That asymmetry is what grinds his gears.

First-class citizens and the hard question

I added my own pet theory here: that a first-class Bitcoin citizen both validates and mines. When Satoshi wrote about nodes in the white paper, he meant mining nodes, because he assumed every user was a miner. Somewhere along the way we amputated mining from validating and convinced ourselves that a node on a shelf (or worse, a node never even used to check your own transactions) confers sovereignty. It does not.

Fifty-one percent of hash rate can activate a soft fork and there is nothing your node can do about it, which is how Bitcoin has always worked. Miners hold governance power most of them never think about, though Kristian offered the charitable and probably correct explanation: mining is brutal, uptime is an art and a daily war, and after the war most operators have nothing left for Twitter governance fights. They outsource the thinking to pools they trust, which loops back to why pool culture matters.

Which brought me to the hard question I promised: Braiins had the first-mover advantage of the entire industry, so why is it not number one in hash rate? His answer was a genuine piece of industry analysis. Pools stopped being a viable core business about three or four years ago, when the FPPS price war hit its floor and Foundry rose. There is nothing left to differentiate a pool on except fee, so the winners were those who treated the pool as a user base to upsell.

Foundry went horizontal: slash pool fees to near zero (0.1%, numbers no one can compete with), then make money on hosting, sourcing, and insurance sold to the biggest publicly traded miners, just as the hash rate migrated from China to the US. Braiins went vertical: keep a sane pool margin and build firmware and management software on top, serving mid-size miners who are politically aligned with Bitcoin. The kicker, which Kristian flagged as possibly not public information, is that Foundry has reportedly been raising its pool fees drastically over the past year because the model is struggling.

Meanwhile Braiins is a 120-person company that has to make payroll and, in his words, is capitalist, not communist. It fielded acquisition offers from public companies during bull markets and always refused, because the founders simply like building things. His closing line deserves the headline it will not get: when was the last time Foundry did something cool? You will not remember, because they do nothing. They mine blocks.

The giveaway, and how to enter

This episode’s giveaway is a hard-mode one, because the prize deserves someone who actually wants it: a Braiins Dock (the $300 touchscreen smart display with price alerts, graphs, and pool monitoring, programmable, and yes, it runs Doom, a steal compared to the $500 clock that only shows the price), a copy of Kristian’s book, and $21 of hash rate on the marketplace to try your luck. The mechanics: subscribe to the Braiins newsletter at braiins.com/newsletter, subscribe to this podcast, and follow Kristian on X. The winner gets picked from newsletter signups dated after this episode and must produce screenshots of the other two. He is a marketing guy; he has KPIs to hit. I respect the honesty.

This episode is sponsored by Cake Wallet, LayerTwo Labs, Bitcoin.com News, SideShift.ai, and Braiins Hashpower!

Cake Wallet funded the one Zcash giveaway (about $204 at recording time), which Kristian received live on air and pledged to immediately swap to Bitcoin and then gamble on rented hash rate, a full sponsor economy in one transaction. LayerTwo Labs keeps building drivechains; as established above, even the Braiins Superman likes BIP 300. Bitcoin.com News has completed its orange rebrand and finally learned the difference between BCH and BTC, with a nicely balanced institutional-plus-political mix. And SideShift remains the place to swap whatever stablecoin or CBDC wannabe someone paid you into real Bitcoin, straight to your wallet, no account required.

Listen to Kristian on YouTube, Spotify, Apple Podcasts & More!

Watch on YouTube

Time stamps:

00:01:09 – Welcome: S17 E14 with Kristian Csepcsar, Chief Meme Officer of Braiins
00:01:45 – The Bitcoin inflatable (or giant cheese roll)
00:02:13 – Recording in person at the undisclosed Braiins office
00:02:58 – Is Braiins hardware, software, firmware, pool or books? Yes
00:03:06 – The Ultimate Bitcoin Mining Guide: free download & audiobooks
00:03:19 – Only two people ever pronounced “Csepcsar” correctly
00:04:25 – A nonprofit book: $10 on Amazon, cheaper than your toner
00:05:15 – Guy Swann narrates the audiobook
00:05:22 – A gift from Norway: whale salami
00:05:55 – GreenTech: a 2 MW mining farm heating an office past the Arctic Circle
00:06:31 – Vlad’s two years of vegetarianism
00:07:02 – “It’s not made from the whale penis, right?”
00:07:42 – 1.3 million BTC mined: the salami as a flex
00:08:04 – Why it is called Slush Pool: Bitcoin.cz & the email signature
00:09:53 – The oldest Bitcoin company ever
00:10:09 – “There is no I in Braiins” (there are two)
00:10:32 – The plan: a short history of Bitcoin mining
00:10:53 – Sponsors & the Zcash giveaway: Braiins mined Zcash early
00:11:31 – LayerTwo Labs & BIP 300: very good for the miners
00:11:46 – Bitcoin.com News learned the difference between BCH and BTC
00:12:26 – Vlad’s airdrop: installing Cake Wallet live
00:12:59 – The Braiins Dock giveaway: a $300 smart display that runs Doom
00:13:41 – The $500 Blockclock comparison
00:14:08 – SideShift: swap the CBDC wannabes for Bitcoin
00:14:34 – Sending one Zcash live on air ($204)
00:15:27 – The plan: swap to BTC & buy hash rate
00:15:38 – Braiins Hashpower: on-demand SHA-256 hash rate
00:16:09 – Solo mining Bitcoin Cash & a depressing observation
00:16:40 – Hash price at all-time lows & the hash rate that won’t turn off
00:17:33 – Solo mining is a lottery, but better odds than Powerball
00:18:12 – The powerbomb wrestling interlude
00:18:27 – Rolling the dice every ten minutes
00:19:19 – A solo block hit in the first month of public beta
00:20:14 – The B2B use case: covering contractual uptime
00:20:47 – Another block found on Public Pool
00:21:13 – The best marketing that money cannot buy
00:22:05 – When mining farms pivot to AI
00:23:06 – BitAxe money buys 1,000x the hash rate
00:23:31 – The $400 BitAxe with the terrible fan
00:23:57 – “We just came”: the chief of commercial returns
00:24:17 – 9 out of 10 public miners are pivoting to AI
00:25:18 – The 2017 blockchain name-change déjà vu
00:25:35 – The giveaway challenge: newsletter, subscription & screenshots
00:27:06 – Adding $21 of hash power to the prize
00:27:30 – What $6 buys: one petahash for two hours
00:27:52 – The Braiins mini miner & promo code BTCTKVR
00:28:03 – Dual mining BTC & BCH: two lottery tickets at once
00:28:37 – How much to 51% attack BSV? A few dollars
00:29:10 – The 5% pool that mined 1.3 million bitcoin
00:29:46 – One CPU, one vote & Satoshi’s naivety
00:30:18 – Satoshi’s email to Laszlo Hanyecz: please stop GPU mining
00:30:51 – Zooko’s internal email: how Slush discovered Bitcoin
00:31:11 – The birth of pooled mining
00:32:07 – Why Prague became a Bitcoin citadel
00:32:37 – The first block: December 16, 2010
00:33:25 – Pudding Pop’s pool: the forgotten week-earlier attempt
00:34:15 – The first successful implementation that stood the test of time
00:35:44 – GPUs boom, CPUs die & why pools were invented
00:36:16 – 2013: the ASICs arrive
00:37:22 – Canaan: first to market, worst at business
00:38:05 – Butterfly Labs & the coordinated napkin-math tragedy
00:39:18 – The first halving & the tissue interlude
00:39:29 – Why the pool fee is 2%: an urban legend
00:40:20 – Chinese pools, FPPS & the price war
00:42:11 – Sixteen years of baggage
00:42:33 – The pool’s birthday is in November: party plans
00:43:16 – The 2011 Linode hack: 3,000 BTC gone
00:43:55 – Slush paid everyone from his own pocket
00:44:15 – How the hack created Trezor & brought Braiins in
00:45:18 – No third parties ever again: encrypted in-house hardware
00:45:47 – Running a pool from dates & the opera
00:46:35 – Kristian’s path: from KPMG IT audit to mining memes
00:47:27 – The hash price pain chart
00:48:12 – Turning point one: the 2017 bull market
00:49:23 – Turning point two: the China ban (more important than Taproot)
00:49:30 – How the China ban got Kristian married
00:50:07 – 15% of hash rate is still in China & getting banned again
00:51:07 – Thank you, President Xi: the C.R.E.A.M. tribute
00:52:17 – The most political pool: Bitcoin XT & the DDoS attack
00:53:21 – First pool to signal Taproot
00:53:47 – Bitcoiners first, business second
00:54:47 – BIP 110 (formerly BIP 444) & the Superman on staff
00:55:48 – Lukas’s job title is literally “Braiins Superman”
00:56:17 – Pools as educators for business-minded miners
00:57:12 – Vote with your hash rate
00:58:00 – The 444 vs 110 numbering mystery
00:59:11 – Do pools own the ASICs? A common misconception
01:00:12 – “Mine as one”: how pooling actually works
01:01:17 – Why Braiins never penetrated Russia: the kickback requests
01:02:08 – China, Europe & the Paraguay-Ethiopia trap
01:03:13 – The Houston site: 6 MW to publicly prove the firmware
01:03:51 – Texas: the beating heart of Bitcoin mining
01:04:07 – The pitch for home miners: no KYC
01:05:06 – NiceHash has KYC; Braiins wants an email address
01:06:06 – Lightning payouts: freshly minted beer money
01:07:14 – The S19 XP story: firmware that doesn’t die when hot
01:09:09 – Dynamic power targets: the no-brainer Bitmain never built
01:10:24 – Bitmain’s playbook: slash prices, release, repeat
01:11:11 – The Bitmars ambassador announcement & the angry YouTubers
01:12:38 – Intel, Samsung, Auradine: the challenger graveyard
01:14:19 – What the market needs: a chip supplier, not another Apple
01:15:19 – Bitdeer: can Bitmain’s founder challenge Bitmain?
01:16:55 – Block’s Proto miner: a fan’s skepticism
01:17:22 – Jack Dorsey’s investment record: LDK & the screenless Bitkey
01:19:15 – A short history of hardware wallets with screens
01:19:34 – The Proto feedback incident: when companies attack the critic
01:22:06 – The Coldcard “open source” license debate
01:23:08 – Attack the idea, not the person
01:24:03 – The centralization conversation nobody is having
01:24:42 – Closed-source firmware & the golden nonce thought experiment
01:26:48 – Foreman: the management software monopoly
01:29:52 – Sticky software: two minutes to switch pools, months to switch managers
01:31:14 – First-class Bitcoin citizens: miners & soft forks
01:32:46 – Uptime is an art: why miners outsource governance
01:34:22 – The hard question: why isn’t the oldest pool number one?
01:35:31 – When pools stopped being a core business
01:36:07 – Horizontal vs vertical integration: Foundry vs Braiins
01:38:17 – Sustainable margins over vanity hash rate
01:39:06 – “We’re capitalists, I’m not a communist”
01:39:57 – When was the last time Foundry did something cool?
01:40:13 – Outro: newsletter, merch & discount codes
01:40:38 – Heating a house with one hydro miner: the new video
01:41:05 – Final sponsor thanks & goodbyes

Verify this episode

This is the SHA-256 fingerprint of the MP3 as published here, timestamped in the Bitcoin blockchain with OpenTimestamps. If one byte of the file changes, the fingerprint changes.

SHA-256
787963291a89b00fcacfab59aa8bfcc1f2864db9fe788ad38318e86c66244287
Proof
2026-03-20_s17_e14_kristian_czepczar_on_braiins_the_history_of_bitcoin_mining_final.mp3.ots · all episode hashes (proof)
Status
Anchored in Bitcoin block 974,312
How to check it yourself
  1. Download the MP3 and the proof into the same folder.
  2. Run sha256sum 2026-03-20_s17_e14_kristian_czepczar_on_braiins_the_history_of_bitcoin_mining_final.mp3 (on a Mac: shasum -a 256 2026-03-20_s17_e14_kristian_czepczar_on_braiins_the_history_of_bitcoin_mining_final.mp3). The result must match the SHA-256 above.
  3. Install the OpenTimestamps client (pip install opentimestamps-client) and run ots verify 2026-03-20_s17_e14_kristian_czepczar_on_braiins_the_history_of_bitcoin_mining_final.mp3.ots. It hashes the MP3 again and checks the proof against your own Bitcoin node, then prints the block and the date the file existed by. Without a node, ots info 2026-03-20_s17_e14_kristian_czepczar_on_braiins_the_history_of_bitcoin_mining_final.mp3.ots shows the block height to look up in any block explorer.

Share: X · Email · Link

Related

Comments

Comments run on my own server. No ads, no trackers.

0:000:00