Tak.peaq✓

While most people speculate on #DePIN tokens, I operate the infrastructure behind them. 📡 Connectivity & Network Data – @375ai (network & connectivity data → $EAT) – @AnyoneFDN router (bandwidth sharing → $ANYONE) – @roam_network (WiFi mapping & connectivity validation) 🗺 Mobility & Mapping – @MapMetrics (anonymous travel data → $VAULT → $MMAP) – @captur_go (mapping & geo-data collection) 🧠 Compute – @Acurast phones (mobile compute → $ACU) – @DATSProject node (desktop + mobile) – @KaisarNetwork node – @teneo_protocol AI agent (public data indexing) 💾 Storage – @DeNetPro Storage node / Watcher node 🎙 Data Contribution – @silencioNetwork (noise + voice data → $SLC / $USDC) – @ReflexDAO (health metrics → $BEAT) – @ScannitNetwork (receipt scans & microtasks) 🌍 Environmental & Energy – @airalabs_io air quality sensor – @shelly_IoT plugs + @combinderio (energy data → $BIND) – @SkyX (weather data, soon) 💰 Capital Allocation – RoboFarm RWA investment (monthly yield, tokenized agriculture) – @peaq staking & long-term holding -> $PEAQ I test these systems in the real world. I collect data, rewards, and firsthand experience. And I share honest conclusions — even when they’re inconvenient.

I completed a 100-day quest on @roam_network a few days ago and I was looking forward to the next one... Today @roam_network announced they're shutting down. That hurts. #DePIN is a bet. Sometimes it pays off. You collect data, provide bandwidth, storage or compute and actually earn money. Sometimes you spend months contributing, collecting points and completing quests, hoping that eventually it will lead somewhere. Then it doesn't. @charge_xyz: gone. @ScannitNetwork: gone. @MapMetrics: technically still here, but somehow also not. And now @roam_network joins the list. To be fair, it's not always about money. Climbing a leaderboard can be fun. Completing a long quest feels good. A streak gives you a reason to come back tomorrow. I've spent ridiculous amounts of time doing exactly that, and I enjoyed some of it. But that's also part of the risk we take as DePIN users. We don't only contribute data, bandwidth, storage or compute. We contribute our time and attention. Sometimes we get tokens for it. Sometimes we get a useful product, a community or simply the satisfaction of reaching the top of a leaderboard. And sometimes a project you've spent months contributing to simply disappears. That's DePIN too. And the best part? A few hours after announcing the shutdown, the app prompts me with "Update available." You literally can't make this up.

In case you're lost on Crypto X. What did I miss? Add your translation below. 🔽 (Text: me / Image: AI)

Delivery robots paying for their own electricity. Machines taking out loans. I looked at what the 5 winners of the @peaq / @SuperteamDE hackathon actually built. This is what the Machine Economy looks like. 🥇 @R0boPay: payments for machines RoboPay built a payment system for machines and AI agents in the physical world. Their prototype uses a delivery robot: the payment is set aside before the delivery, the robot completes the job and proves it, and only then is the money released. @solana handles the payment, peaq gives the robot a verifiable digital identity, and Staex provides secure connectivity. The current version already works with a real small delivery robot and a QR code, while additional verification through the letterbox is planned. 🥈 FieldProof by @excelhtf: software buying fresh data from machines FieldProof lets software pay device operators for fresh information from the physical world. A piece of software could ask something as simple as: "Is this door open right now?" A sensor performs a new measurement, signs the result and the operator gets paid. The prototype already works with a real sensor. peaq gives that device an identity and records its work history, so the buyer can verify which registered device produced the reading. Importantly, this proves that the device made the claim, not that the claim itself must be true. 🥉 Fiducia: giving machines spending rules Fiducia explores what happens when machines are allowed to spend money themselves. In its demo, a delivery robot earns money, needs electricity and chooses between different charging stations based on price and track record. But the owner remains in control. The owner decides in advance which providers the robot may use, how much it may spend on one charge and how much it may spend in total. The robot, driving and charging are simulated, but the peaq machine identities and activity records and the Solana transactions are real. 4️⃣ SKIM — credit for machines SKIM asks a fascinating question: if a machine generates revenue, why couldn't it borrow against that revenue? A machine receives an advance. As it continues earning money, part of every payment automatically goes toward repaying that advance. Its digital machine identity acts as security until the money is paid back. And just like with humans or companies, building a good financial track record matters: repay successfully and the machine can qualify for a larger and cheaper loan next time. The team has already run and repaid two test loans on peaq's Agung testnet, although the credit ratings used there are simulated. 5️⃣ @Desc_Protocol: escrow with AI arbitration DeSC tackles another basic problem: how do two parties transact when neither wants to trust the other? USDC is locked in a Solana escrow contract and the work is delivered. An AI moderator chosen beforehand checks the delivered files against the agreed criteria. If the work passes, the contractor gets paid. If it fails, the buyer gets refunded. It's currently running on Solana Devnet, so we're still talking about a prototype rather than a production system. Put these five projects together and you get a pretty good picture of what a Machine Economy could look like. Machines earn money. They spend it. They buy data and services from other machines. They build a financial track record, take out loans and prove that work was actually done. peaqOS provides a foundation for that: digital identities and verifiable history. @peaq isn't here to build every piece of the Machine Economy. It's here to make sure they all work together.

Well... we're back here again. 😅 A few months ago my @AnyoneFDN dashboard suddenly showed 0 registered relays while my Anyone Router was happily running in the background. It turned out to be an indexing issue and was eventually fixed. Now the exact same thing seems to be happening again. My dashboard currently shows: → 0 registered relays → 0 claimed relays → 0 active relays → 0 claimable relay rewards The last time I checked, I had more than 120 $ANYONE waiting to be claimed. Meanwhile, the router itself couldn't look much healthier: relay running, client online, OR-Port reachable and all the nice FAST, STABLE, RUN, V2Dir and GUARD flags are there. So I'm not touching anything. Anyone else seeing this again, @AnyoneFDN?

$SLC: +700% in three months! I know this sounds like an "I told you so" post, but I don't predict prices. I judge projects by their business model, their execution and whether people are actually willing to pay for what they build. So forget the price for a moment and look at the numbers behind @silencioNetwork: ✅ 2.5M+ contributors across 180+ countries ✅ 1M+ hours of off-the-shelf voice data ✅ $250,000 contract with one of the world's leading Voice AI companies ✅ Biggest deal in Silencio's history signed with a frontier AI lab for African voice data ✅ 2026 revenue expected to cross $1M ✅ $40M+ active enterprise pipeline across 50+ leads ✅ 5 of the world's top 10 AI companies have already inquired about Silencio's data And there could be much more ahead. AI models are still hungry for voice data, especially across languages, dialects, accents and regions where high-quality datasets are difficult to find. Once Silencio reaches a $1M+ cash reserve, the company has committed 75% of data-sale revenue to the BlockSound Foundation, which will use it to buy $SLC on the open market and burn it. So if the business grows, there is a direct mechanism designed to turn part of that revenue into buy pressure and permanently reduce supply. AI company needs data → Silencio mobilizes contributors → contributors get paid → Silencio generates revenue → revenue can eventually fund $SLC buybacks and burns. That's a #DePIN model I understand. And judging by the growing commercial interest and the recent price action, it looks like more and more people are starting to understand it too.

Yesterday I wrote that $SLC was up 27% in a week and explained why I like @silencioNetwork. Apparently I need to do that more often. What the hell happened here? 😂

Most of the crypto market is bleeding. Meanwhile base:0x6bd83abc39391af1e24826e90237c4bd3468b5d2 went up 27% in one week. Why? As always: No idea. 😂 But I can give you a few reasons why I think @silencioNetwork is one of the best #DePIN projects. First of all, I love how simple the idea is. You don't need special hardware. You don't need to run a node. You just need a smartphone. Record your voice. Contribute speech data in your language. Get rewarded for it with $USDC and base:0x6bd83abc39391af1e24826e90237c4bd3468b5d2 . And with 2.5M people in the network, that becomes especially interesting for languages that aren't spoken by hundreds of millions of people. Silencio says it collected 100,000+ hours of voice data across 45+ African languages in just eight months. Some of those languages barely have any commercial speech datasets available at all. And the market for that data is real. @thomasmesserer recently pointed out that one Voice AI lab committed hundreds of millions of dollars to collecting data across 22 languages in a single country. That's not money going to Silencio. But it tells you something about how valuable good voice data can become. Does any of that explain base:0x6bd83abc39391af1e24826e90237c4bd3468b5d2 going up this week? No idea. But while most of my portfolio is currently doing its best impression of a falling knife, I'll take it. I only have one complaint: When do we finally get a Swiss German campaign? Gopfetammi! 🇨🇭😂

This is how peaqOS opens doors to a new economy. Literally, this time. @peaq just demonstrated a delivery robot interacting with a building it had never been integrated with before. No custom code for the door. No custom code for the sensors. The robot used its peaqOS identity to discover the devices around it, saw that the loading area was occupied, waited until it was clear, opened the door and drove in. Simple demo. Bigger idea. And that's what I've been noticing with peaqOS lately. Almost every week there's something new. New functions. New integrations. New types of machines using it. New ways for machines to identify themselves, prove things, discover services, interact and eventually transact. Piece by piece, it's starting to look less like a collection of individual features and more like an actual operating layer for machines. I just hope robotics developers are paying attention. Especially the big ones. Because if we're really heading towards an economy with millions or even billions of machines, they can't all live in their own little isolated worlds.

Robots in Korea can now talk to the buildings around them Korea runs 1,220 robots per 10,000 employees, more than anywhere on earth. And the Korea Electronics Technology Institute (KETI) built Mobius, an open platform for a building’s sensors and doors, on the global oneM2M standard. Starting now, a machine running peaqOS can use it on peaq’s Watch a delivery robot read a building it has never seen, wait for its moment, and open the door itself. → peaqOS provides identity, discovery, and coordination → KETI’s Mobius speaks for the building

If you're staking $PEAQ through the @NATIXNetwork validator, check your staking position. I just realized that I haven't received any staking rewards for quite some time. I've had 50,000 $PEAQ staked with the NATIX validator, and everything looked normal in the staking dashboard. But my rewards weren't moving. So I asked around, and @0xEl9ar9ni checked the blockchain for me. NATIX apparently last authored a block on September 5. And since validators and their stakers only receive rewards when the validator authors blocks, that explains why I haven't been receiving any. I've informed @peaq about it as well and I'm changing my validator now. If you're also staking with NATIX, you might want to check yours. I'm posting this because I had no idea anything was wrong. From my side, the staking dashboard still looked completely normal.

“With Web 3.0, we no longer need a middleman. It's time we cut him out.” I still remember that line from one of @peaq's videos more than two years ago. It was one of the reasons I started paying attention to @peaq – and to Web3 in general. And the idea hasn't really changed. But the ambition around it clearly has. @MartinElKhouri recently described where peaq wants to go: “The goal is to be the de facto standard. The financial standard for the Machine Economy. That is what we're building it for.” From cutting out middlemen between people and machines... to trying to build the financial standard for an economy of potentially billions of machines. Same idea. Much bigger scale. And more than two years later, I'm still part of it. On a tiny scale, sure. But it feels good to contribute my little piece.

2.5M+ machines are already online on And the number keeps growing. Some of them move around as robots. Others might be sitting on a desk in your home. That's something I think is easy to miss when we talk about the Machine Economy. “Machine” doesn't just mean humanoid robot. As @MartinElKhouri explained on the @CryptoCoinShow, it can be sensor networks, drones, cars, airplanes, humanoid robots, service robots or collaborative robots. That's a pretty wide definition. A small sensor can be a machine. A smartphone providing compute can be a machine. A robot working in a factory can be a machine. And suddenly, Martin saying that millions are a small number and @peaq wants to eventually reach billions of machines sounds a lot less crazy. Some of those machines might already be in your house. I count 6 smartphones in my office and they will all be part of this economy. That’s the scale @peaq is building for: not a world with billions of robots, but a world where billions of machines can participate in the economy.

At what point does dedication become stupidity? And when does it become money? “I admire your dedication bro.” This comment by @mgt_felix under my latest @Roam_network post is a nice compliment. But it also made me think about how much time, money and hardware I've actually put into #DePIN over the past couple of years. Phones. Routers. Sensors. Smart plugs. Storage. Bandwidth. Compute. Data. And probably thousands of hours of testing apps, completing quests and collecting points. And I'm starting to see a pretty clear divide. Some projects have turned that participation into something real. With @silencioNetwork, I contribute voice data that can be used by Voice AI companies, and I've been rewarded in $SLC and $USDC. Quite a decent amount if you put in the time. With @Acurast, my old smartphones provide compute and earn $ACU. My @AnyoneFDN router provides bandwidth to a privacy-focused routing network and earns $ANYONE. 5 a day, which is pretty nice. My @DeNetPro node provides storage and earns $TBY, although the amount is still tiny. My @375ai_ app earns $EAT, but the fees to claim my token are higher than the actual value of the tokens. With @airalabs_io, I carry an air quality sensor around with me. There are no points or tokens to collect and I don't earn anything from it, but the team keeps shipping updates and developing the product. And then there's the other side of my #DePIN portfolio: Points. Lots and lots of points. Points at @DATSProject or @roam_network. Points and Fragments at @teneo_protocol. XP, $MMAP and $VAULT at @MapMetrics, which seems dead. $NANO at @SkyX_Network or $BEAT from @ReflexDAO, both of which have been silent on X for several months. Quests, streaks and rewards that may or may not become something eventually. And yes, I know there are other DePIN projects out there that are delivering. I'm not saying every early-stage #DePIN needs to generate meaningful income immediately. But after spending this much time actually using these products, I've become much more interested in very simple questions: What is the value loop? What am I providing? Who needs it? Who pays for it? And how does some of that value eventually make its way back to the people providing the resource? “Keep collecting points and wait” gets less convincing the longer you've been doing this. Maybe that's what dedication teaches you eventually. 😂

“That's alpha actually…” @MartinElKhouri just dropped a pretty interesting detail about @peaq's Machine Money Markets in an interview with @CryptoCoinShow. Internally, they're talking about four different rails. @machinedotfun is one of them. Two others are large-scale / industrial financing products that Martin says are already being tested. And the fourth is focused on yield products that could give hardware and #DePIN builders new ways to access capital. That's also being piloted. The plan? “We are planning to roll all of them out within the next quarter. Hopefully.” He also added that rolling them out at scale will probably take longer. So isn't the Machine Money Market strategy. It's one piece of a much bigger one. Market sucks. Builders keep building.

I know I didn't lose any money. But as a #DePIN nerd, this hurts. A few days ago I completed the “Centurion” quest on my @Roam_network app: Scan for 100 days in a row with at least 500 data points each day. Done. 100/100. 🔥 So naturally I looked at the next challenge: “Year of Roam” - do the same thing for 365 days. I was already almost a third of the way there. Or so I thought. The quest started me back at zero. I'm now at 5/365. Meanwhile, the same app proudly tells me: Current streak: 113 days. I haven't stopped scanning. Even after claiming Centurion on October 4, I kept going every single day. So somewhere in the Roam universe, I apparently have both a 113-day streak and a 5-day streak at the same time. No money lost. There isn't even a token with a market value yet. But still. Broken streaks are like relationships ending. You know life goes on. But you still want your 108 days back.

“I think millions is a small number for what we should be able to achieve.” That's what @MartinElKhouri said in his interview with @CryptoCoinShow when asked about the 2.5M+ machines already visible on @peaq's public dashboard on And apparently, that number matters a lot internally: “It's the number one metric that matters. The number one metric that we look at every morning when we open our computer, every night when we go to sleep, is that machine count.” 2.5M+ machines sounds like a lot. For peaq, apparently, it's just the beginning. “The goal is to get it to the billions.” That's quite a target. Ship it! 🚀

Should you post something this embarrassing? Yes, you should. And dear @DeNetPro team: I'm sorry. 😂 For quite some time, I've been trying to get my DeNode Manager to use 9 TB of my 10 TB hard drive. Until now, it was only using the 1,121 GB that had been allocated automatically. It wouldn't work. I updated the software. Changed settings. Sent logs and exchanged messages with support. Tried again. And again. Today, support asked me to record a video showing exactly what happens when I try to set the storage to 9000 GB. So I did. 9000 GB. "Next". Check the other settings. "Save". Didn't work. Then I tried it one more time. 9000 GB. "Save". No "Next". And. It. F*****g. Worked. Apparently, all this time I assumed I could go through the configuration with "Next" and save everything at the end. Nope. The node now has its 9 TB. My dignity has considerably less storage available. Sorry, @DeNetPro. And thanks for your patience. 🙈🙈

Okay, let me explain this new @machinedotfun feature in simple terms. Imagine you're launching a robotics project called $ROBOTOKEN. Normally, you might launch it against $SOL or $USDC. On you can now choose something that actually matches the industry you're building in. How does it work? Let's say you choose solana:Xsc9qvGR1efVDFGLrVsmkzv3qi45LTBjeUKSPmx9qEh, a tokenized asset tracking NVIDIA stock. So your trading pair becomes: $ROBOTOKEN / solana:Xsc9qvGR1efVDFGLrVsmkzv3qi45LTBjeUKSPmx9qEh People start trading $ROBOTOKEN. Trading generates fees. And the creator's share of those fees accumulates in the asset they chose, in this case, solana:Xsc9qvGR1efVDFGLrVsmkzv3qi45LTBjeUKSPmx9qEh. So: People trade $ROBOTOKEN → fees are generated → the project accumulates NVIDIA exposure. If you're building something heavily dependent on GPUs and Physical AI, that actually makes a lot of sense. Building humanoids? You could choose $TSLAx. Building memory-heavy autonomous systems? $MU. And it doesn't have to be tokenized equities. also supports ecosystem assets like $PEAQ, the native token of @peaq's Machine Economy network, as well as solana:69LjZUUzxj3Cb3Fxeo1X4QpYEQTboApkhXTysPpbpump. So a robotics project building around the Machine Economy could choose $PEAQ as its quote asset and accumulate its creator fees in $PEAQ. And of course, $SOL, $USDC and $USDT are still options. That's basically it. Instead of simply collecting your creator fees in $SOL or $USDC, you can collect them in an asset connected to the industry you're building in. What's the opportunity? Instead of collecting fees in $USDC, a project can build exposure to the industry it's betting on anyway. If that industry grows, the value of those accumulated assets could grow too. The risk? You're also adding market exposure. solana:Xsc9qvGR1efVDFGLrVsmkzv3qi45LTBjeUKSPmx9qEh, $TSLAx or $PEAQ can go up. They can also go down. More upside. More risk. Your choice. The takeaway? Your token generates trading activity. Your fees build exposure to the industry you're building in. Pretty clever, actually.

Launch your robotics token against the industry you're building A robotics project depends on a long chain of companies: chips, memory, cloud compute, software and manufacturing. On you can launch your token against one of them, instead of being limited to SOL or a stablecoin. The creator's share of trading fees builds up in that quote asset. Launch against $NVDAx and the fees from people trading your token become NVIDIA exposure. A team building memory-heavy systems could pick $MU, and a humanoid project could pick $TSLAx. Equity pairs on are available through and powered by @Sunrise. Equities: - $BOT (RoboStrategy): Humanoids Index - $MU (Micron): memory and storage for AI and autonomous systems - $NBIS (Nebius): AI cloud infrastructure - $NVDAx (NVIDIA): GPUs, simulation and physical AI - $SPCX (SpaceX): engineering, manufacturing and autonomous systems - $SPYx (S&P 500): broad exposure to the US economy - $TSLAx (Tesla): automation and humanoid robotics through Optimus Ecosystem tokens: - $PEAQ (peaq): Machine Money Markets & robots as financial actors - $CODEC (Codec Flow): Robotics Execution Layer SOL, USDC and USDT are also supported. Not every founder wants equity exposure, so the choice belongs to the project. Doubling down on your industry 1. Your project benefits from the trading activity around its own token. 2. The assets you collect through trading fees may appreciate as the robotics industry grows. Those assets can also lose value, so this is aligned exposure, not guaranteed returns. Build the machine, launch the token, and collect exposure to the industry you're helping grow.

peaq is taking the Machine Economy into a Bundesliga stadium. ⚽ And it's not some small test venue. It's Deutsche Bank Park. Home of @Eintracht. 60,100 seats. 42 hectares. @peaq and @Eintracht are partnering through EintrachtTech for 12 months to explore Machine Economy applications across the stadium and its potentially thousands of machines and connected systems. And some of the ideas they're exploring are pretty wild: ➡️ Cleaning robots paid per completed job through peaqOS Monetize. ➡️ Kiosks, beverage systems and charging points autonomously paying and getting paid. ➡️ Cameras and sensors turning verified data into a product through peaqOS Stream. ➡️ Machine activity and revenue potentially enabling asset-backed financing through peaqOS Verify. And Deutsche Bank Park is supposed to be the starting point. The approach could eventually be adapted to other stadiums, airports, conference centers and campuses. @Ryanbabel, are you listening? 😏⚽

The Machine Economy: expectation vs. reality. 6 @Acurast phones. 1,265.79 $ACU earned so far. And now they're coming to @peaq. The decentralized compute part works. The cable management part is still on the roadmap