Bitcoin Home Mining Playbook with Exergy

↓ Jump to the video and timestamps

I've been stacking hash rate for almost a decade now. It is masochistic. It is also completely addicting. And for years I've watched this idea (Bitcoin miners as home heating infrastructure, as solar monetization tools, as cypherpunk sat-stacking machines sitting in your basement) get dismissed as a niche hobbyist thing. I've never bought that framing. The case was always there.

What was missing was the playbook.

Tyler Stevens and Dylan Seib of Exergy built that playbook, and I partnered with them to get it out into the world. We sat down to walk through the whole thing on tape: the useful miner framework, how to calculate whether your home or business is actually a good candidate, what the building brain looks like and how to build one, hardware selection in a market where S19s are going for around $50, why the AI hyperscaler displacement of on-grid miners is the best thing that's happened to home mining in years, and how the bitcoin actually gets into your wallet when all of this is running.

This is not a "might be worth exploring" conversation. The window is open right now. Here is the map.

Key takeaways

• You don't need to mine profitably. You need to be useful. Every watt you put into a Bitcoin miner comes out as heat. If that heat is cheaper than propane or electric resistance, or if it monetizes excess solar instead of selling it back to the grid for pennies, the miner is earning its keep regardless of hash price.
• The duty cycle is the go/no-go gauge. Cold climates with long heating seasons are the best candidates. The longer the miner runs, the faster the upfront cost pays back. Miners call it uptime; HVAC calls it duty cycle. Same thing.
• AI hyperscalers are handing home miners a gift. S19s are trading near $50 as mega-miners get priced off the grid by AI compute demand. The best hardware entry point in years is open right now, and it won't stay open forever.
• The building brain changes the game. A local Home Assistant instance running on a Raspberry Pi can dynamically switch between gas heat, Bitcoin mining, and solar monetization in real time, and an AI like Claude can write the automations for you once the hardware is rigged up.
• Don't rip out your furnace. Size the miner for your average heat load, not your peak cold day. Let your existing furnace handle the single-digit nights. The hybrid approach keeps upfront costs low and duty cycle high.
• Stack sats non-KYC from your own home. Point your miner at Ocean with your own Bitcoin address, add a Lightning payout route via Bolt 12 and Coinos or a Core Lightning node, and the subsidy lands directly in your wallet. No exchange, no KYC, immediately spendable.

You have an advantage. Use it.

The first mental shift is understanding that a home miner is not competing with Riot Platforms or CleanSpark. It is competing with your propane bill. That reframe changes everything.

Tyler's framework for this is the useful miner. Every building on earth that is heated has a sunk cost on that heat. The physics are simple: all the electricity you put into a Bitcoin miner comes out as heat. So the question is not "can I mine profitably at residential electricity rates?" The question is "can this machine be useful?" Useful means one of three things: it heats your building for less than your current fuel costs when you count the bitcoin back as a rebate; it monetizes excess solar generation instead of selling it back to the grid for airline-miles-style credits; or it mines opportunistically when hash price rips and the numbers just work.

You want the tool in the building so you can do all three.

This is why I've always found home mining appealing beyond the pure economics. It's cypherpunk in the real sense. You're stacking sats in a non-KYC fashion while subsidizing infrastructure you'd be paying for anyway. As Tyler put it on tape, the best mining operation is one that isn't seeking profitability in the traditional sense. The sunk-cost miner at home, running against a heating load that exists regardless, is structurally positioned better than a datacenter chasing the margin.

The calculus comes down to two things: is your current heating fuel expensive enough that a miner with bitcoin rebate can beat it on a per-kilowatt-hour equivalent basis, and is your heating season long enough that the miner will run enough to justify the upfront cost? Both are answerable with your utility bills, the Home Mining Playbook, and the energy arbitrage calculator on Exergy's site.

The fuel comparison: is your home a good candidate?

Dylan's starting point is always: what are you heating with right now? Most people don't think about this carefully, but it is the single most important variable in whether home mining makes sense for you.

In the US, the options are natural gas, propane, heat pumps, electric resistance, and wood. Natural gas is the hardest case to beat because it is cheap and abundant in most of the country. If you have cheap piped gas and expensive electricity, the economics are tighter. Propane is a different story.

Propane is expensive and variable, delivered in bulk at prices that swing seasonally. Dylan's napkin math: propane can work out to around 9 cents per kilowatt-hour equivalent, while running an electric miner at 10 cents per kilowatt-hour, with 3 to 5 cents back in sats, nets out to a 40 to 50% reduction in effective heating cost compared to propane. That is one customer scenario from their calculator, not a guarantee for every setup, but it illustrates the range.

Already on electric heat? That's the easiest switch of all. You're just replacing one electric heating element with another that also produces bitcoin.

Solar changes the math in your favor even against natural gas. If you have a grid-tied solar array, your excess generation gets sold back to the grid for pennies (literally worse than airline miles, as Tyler put it). Blow that excess through a miner instead and you are capturing real value. The 3.3x multiplier on solar monetization that came up in the data Dylan shared during our screen share is one customer's result, but the direction is consistently favorable for anyone with meaningful solar surplus.

The second variable is the duty cycle. Tyler translated it from mining language: uptime. How often is the miner actually running? If you live somewhere cold for months (not just one brutal week), the miner runs long enough to pay back its upfront cost and then some.

Denver in January is different from Phoenix. The colder and longer your heating season, the stronger your candidacy.

The building brain: how smart control makes it work

Once you have miners in a building, you need a way to coordinate them with everything else happening in that building. That coordination layer is what Tyler and Dylan call the building brain. It is the most important piece of the system and the one most people have never thought about.

Tyler uses a hybrid car analogy to explain it. A RAV4 hybrid doesn't have two full-size powertrains. It has a small battery, a small electric motor, and a gas engine, and the car's brain decides in real time which to use and when. The result is dramatically better efficiency than either system alone.

The building brain works the same way: it knows your thermostat call-for-heat, your solar inverter's current output, the bitcoin hash price, your electricity rate (including time-of-use changes), and the cost of your gas. It runs all of that in real time and decides which fuel source is cheapest at any given moment.

Dylan built this on Home Assistant, an open-source, locally-run home automation platform that runs on a Raspberry Pi. Locally run matters. The building brain should live in your building, not on Ring's cloud server or Google Nest's subscription infrastructure. You own the logic, you control the data, and it physically connects to your thermostats, your solar inverter, and your miners.

The automations are where it gets interesting. Dylan's own system screamed at him during a recent 10% downward difficulty adjustment: difficulty dropped, hash value was excellent, he had solar running, so the brain said crank all the miners, dump the heat outside, and stack. One of their customers set the logic differently: prioritize SAT stacking over gas always. Even in moments when gas heat is technically cheaper, his building chooses the miner first because he wants the sats.

The set-and-forget goal is real. You tell the brain your preferences once and it optimizes from there.

The barrier to building this is lower than people think. Tyler was on Claude, fed it a draft of the playbook, had the Home Assistant MCP connector running, and asked it to look at all the sensors in his house, find the Bitcoin miner, and build the automations for whichever scenario he wanted. Done.

This is the point I don't think we can overstate: the intellectual capital required to set this up is collapsing in real time, at exactly the same moment the hardware is becoming affordable. These two adoption curves are running up and to the right together.

Hardware selection: the right miner for the job

The hardware question depends entirely on your goal. That is the first thing Dylan asks every person who reaches out to Exergy.

If you want to tiptoe in, start with space-heater-class miners. The Avalon Mini 3, the Stealth miner, and the Slim 19 are all in the roughly 850-watt range. They are small, modular, easy to place in individual rooms, and require no major electrical work. Plug one in, pair it with Home Assistant via the miner integration, add a digital thermostat you can control from your phone, and you have a working building brain starter kit.

Your wife may still object to the noise, but that's a separate problem.

For baseload integration into an actual furnace or boiler system, the S19 is the obvious choice right now given pricing. Tyler's example: if your building needs 5,000 watts of baseload heat, you could buy one new 5,000-watt miner for roughly $3,000, or you could buy two S19s at around $50 each and have 6,000 watts of heat-producing hashrate for under $200. The sizing philosophy here is important: you are not sizing for the coldest day of the year. You are sizing for your average heat load. Let the existing furnace handle the brutal single-digit nights.

That keeps your upfront cost reasonable and your duty cycle high.

Dylan's Denver example makes this concrete: on average winter days in the mid-20s to 30°F range, a couple of Mini 3s handle the load. When it drops to single digits, the furnace steps in. That's the hybrid car approach applied to a real home.

Firmware and control capability are underrated selection criteria. Most miners ship with closed-source firmware from Chinese manufacturers. Some have aftermarket options (Braiins and LuxOS are the main ones) that play nicer with Home Assistant and give you real API access. Without that control layer, integrating the miner into a smart building system is fighting against the hardware instead of with it.

One more hardware reality worth stating plainly: Bitmain's latest machines are three-phase hydro miners. Residential homes in North America run on single-phase power. For now, S19s and earlier single-phase hardware are the only practical residential option.

That is also why the $50 S19 window matters. Use it while it's open.

The mega-miner apocalypse is your opportunity

The displacement of on-grid Bitcoin miners by AI hyperscalers is the most important structural shift happening in mining right now, and it is unambiguously good news for home miners.

One executive Tyler cited put it at 60x dollars per megawatt hour. AI compute is paying roughly 60 times what Bitcoin mining pays for the same electricity. For large publicly traded miners with shareholders and fiduciary obligations, the answer is obvious. Pivot. And they are. CleanSpark, Riot, Core Scientific are all moving toward AI hosting and HPC. That is their job when the economics are that lopsided.

Dylan's read on this is exactly right: every announcement of a mega-miner pivoting to AI is music to his ears. Network hash rate drops. Difficulty adjusts down. Home miners collect more sats per kilowatt-hour. The network's defensive moat (the fact that it is financially irrational to attack it) actually strengthens as mining becomes more distributed.

A thousand homes each running two S19s is a more resilient network than ten datacenters running a hundred thousand machines. Physical decentralization is the long game here.

The hardware flood is already starting. S19s are trading at around $50 to $100 right now as industrial operators liquidate. I've been waiting for this moment for years. The arc of mining history has gone CPU to GPU to FPGA to ASIC to industrial scale, and now it's bending back toward the individual.

I don't know exactly what the end form factor looks like, but I'm pretty confident mining ends up as a collaborative part of a home brain, helping with heating, taking advantage of excess solar, firing up when the hash price is right.

The longer-term infrastructure question is real. Tyler flagged it on tape: if Bitmain stops selling to industrial miners and shifts entirely to three-phase hydro hardware, they eventually need to build something suitable for residential and commercial building integration. The current industrial form factor is not it. The open-source firmware stack being built at the 256 Foundation (Mujina firmware, with Ryan as lead architect) is the necessary infrastructure to get there.

Think of it as the Linux of mining firmware: open-source, inspectable, auditable, running on multiple chip architectures. It runs on S19s, it runs on Bitaxes, and it's being designed to run on 256 Foundation's open hardware including LibreBOARD, which lets you connect hash boards via USB-C and mix and match boards from different machines. The dev calls are open. If you care about how home mining firmware should handle solar dispatch or heating control, get in there and say so.

Getting paid: how the bitcoin actually reaches your wallet

Dylan told a story from an event they hosted: someone asked, "Okay, so how do I get the bitcoin out of the miner?" It stuck with him. The files are in the computer.

The answer is pool mining, and Dylan's recommendation is Ocean. Ocean lets you build your own block template and receive payouts directly to a Bitcoin address you control. You provide your address, the pool pays it out on-chain when you hit the minimum threshold (verify the current payout minimum on Ocean's docs before you set this up, as it changes). No exchange, no KYC, no intermediary holding your sats.

If you're running lower hash rate and don't want to wait for the on-chain minimum to accumulate, Ocean supports Lightning payouts via Bolt 12. Set up a Core Lightning node or use Coinos with a Bolt 12 offer linked to your Ocean account and sats flow to your Lightning wallet as you earn them.

This is the part that beats the hell out of credit card rewards. Your subsidy for running infrastructure in your home is immediately spendable bitcoin, landing in a wallet you control. Hold it, sweep it to cold storage with a Bitkey or another hardware wallet, spend it. Your call.

No one else touched it on the way there.

Tyler also flagged the longer-term vision here: decentralized pool protocols that remove the pool operator as a coordination point entirely, using a shared chain architecture. Tyler has been working through it with Matt Corallo and one of the 256 Foundation's engineers, alongside Stratum V2. It's early, but the direction matters.

Physical decentralization of hash rate into homes is one half of the equation. Digital decentralization of pool coordination is the other.

How to get started, and when to call Exergy

Dylan's starting checklist is simple. Have your goal in mind: heating, solar monetization, sovereignty, or some combination. Pull your utility bills. Read the Home Mining Playbook and run the energy arbitrage calculator on Exergy's site. Decide how integrated you want to go.

Space heater tiptoeing or fully plumbed into a furnace or boiler.

Tyler's tiptoeing path if you already have a miner or just picked up a $50 S19: grab a Raspberry Pi and install Home Assistant (it's also packaged on Umbrel and Start9 if you're already running a node). Download the miner integration from the Home Assistant community store. Watch a few YouTube videos, attend Dylan's weekly Home Assistant office hours, read the Exergy documentation.

Then connect your Claude or OpenAI account to Home Assistant via the MCP connector and start asking it to build your automations. The barrier is lower than it looks once the hardware is physically in place.

Do not reinvent the wheel. Tyler and Dylan have already done the hard work of figuring out what works and what doesn't, what sizing logic holds up in real installs, and what integrations actually function. I am a don't-reinvent-the-wheel guy. Use what they've built.

For TFTC roundtable members: we have a closed Zoom session with Tyler and Dylan scheduled for July 14th to go deeper on this. Your questions, your specific setups, more detail than we can cover in a podcast. Sign up for the roundtable at tftc.io if you want to be in that room.

For everyone else: reach out to Exergy at exergyheat.com. You can do this yourself with what's in the playbook and on their support page. But if you want to move faster and get it right the first time, talking to Tyler and Dylan is the fastest path.

Dylan's closing line is the most honest thing said on this episode: stacking hash rate is more addictive than stacking sats. I co-sign that completely.

About Tyler Stevens and Dylan Seib

Tyler Stevens and Dylan Seib are the team behind Exergy, a company focused on home Bitcoin mining and hashrate heating: integrating ASIC miners into residential and commercial heating systems, solar setups, and building automation. They have spent the last two years focused on this space, building Home Assistant integrations for miners, running real installs, and publishing their findings. Together with TFTC they authored the Bitcoin Home Mining Playbook, and Dylan hosts weekly Home Assistant office hours for people building their own setups. Find them on X: Tyler and Dylan.

Sources mentioned

• The Bitcoin Home Mining Playbook (TFTC x Exergy): the playbook this whole conversation walks through
• Exergy: Tyler and Dylan's company, home of the energy arbitrage calculator and install documentation
• Home Assistant: the open-source, locally-run automation platform behind the building brain
• Ocean: the pool Dylan recommends for direct, non-KYC payouts to an address you control
• 256 Foundation: the nonprofit building open-source mining infrastructure, including the Mujina firmware and Libre Board
• Stratum V2: the open mining protocol work running alongside the decentralized pool efforts discussed

Watch the conversation

Timestamps

• 0:00 - Intro: money freer than free
• 2:05 - Two years down the hashrate heating rabbit hole
• 8:37 - BTUs, fuels, and the heating cost math
• 13:47 - Solar monetization as the second use case
• 31:05 - The building brain: Home Assistant
• 35:02 - Hardware selection: what's your goal?
• 46:02 - Sizing for average load and the 45% savings example
• 53:32 - The 3.3x solar multiplier
• 1:01:16 - The mega-miner apocalypse
• 1:05:55 - Mujina, Libre Board, and open firmware
• 1:14:38 - Getting started with Home Assistant
• 1:16:06 - How the bitcoin reaches your wallet
• 1:19:43 - The playbook, the Round Table call, and wrap

Sponsors

• Bitkey: Bitcoin self-custody hardware with no seed phrases, two-of-three multisig, and a full screen so you can verify what you're approving, with inheritance recovery built in. Code TFTC for 10% off. bitkey.world
• CrowdHealth: Crowdfunded healthcare that Marty and his family of five have used for five years. You pay the first $500 on regular health events, the first $3,000 on a pregnancy, and the crowd covers the rest. Code TFTC for $99/month for the first three months. joincrowdhealth.com/tftc
• Lygos Finance: Non-custodial lending on Bitcoin's base layer. Your keys, your collateral, verifiable on-chain. Tell them TFTC sent you.
• Salt of the Earth: Pink Himalayan salt electrolyte mix, no sugar. Code TFTC for 15% off. drinksote.com/tftc