Thank you to everyone who’s been following along🥹 Just got the Influencer tag and honestly did not expect it. Grateful for every follow, reply and conversation along the way. Appreciate you all & let’s keep Building....🥰

A buyer in munich asked for the isar steps wet from Rain, at Dusk, Puddles and Dark corners included. that brief went up on @vangrid_io yesterday for $1,410 and was accepted about 7 hours later. the ask was every step from the top terrace down to the water, the stone edges, the railings end to end,and the waterline at the foot of the wall. two or three evenings were fine if one visit was not enough. a newer one in savannah reads the same way. a reefer block at a container Terminal, both racks end to end, plug points and cable trays, Recorded at Night with only the yard Lighting on. no extra Lamps, because the lighting is what they want to see. $1.5k in Usdc, still open. The hour and the weather are part of the order now, and someone who lives nearby is already out at dusk.

Most launches ask you to show up early. this one asks what you already did. zETH registration opens today on @zerufinance ,the first zToken built from Ethereum activity and run as a beta. getting ready is simple, sign in to Zaps, link the wallets you really use, and mint your zScore, since you need a minted zScore to win a seat in the race. the interesting part is what counts. paying fees, paying interest, supplying capital. wash trades and trading the market's own token score nothing, so the board leans toward people who were using Ethereum anyway. the community also picks the coin's name and art before the race starts,and more lifetime zaps give more votes, by square root. The seat goes to the wallet that did the work. Enter: app.zaps.wtf

a builders hoarding panel is the most Ignored thing on any Street, and someone just posted a paid series for Seven of them. it's a bulk order on @vangrid_io. seven separate bounties, $100 each, same brief every time,one panel face on, the joint to its Neighbour, and the ground line along the bottom. five of them were accepted on oct 24, each one inside half a day of going Up. Item 3 was still open this Morning. think of the site you pass on the way to work. one short steady walk past a panel with your phone, and the buyer gets another version of the same Obstacle in a Different place. seven slightly different panels probably teach a robot more than one perfect one. there have been other series too. bollards in ghent, parking meters in raleigh,stair steps in cambridge, utility boxes on poles in cologne. vangrid's own stats count 200 bulk Bounties so far, worth $22,120. i like this format. the street stuff nobody looks at turns into small paid jobs for whoever lives nearby.

One of the latest trades on the Ditto feed this morning was a $39.90 SOL long, and the trader behind it is sitting at rank 15. That's the bit people miss about the $50K Ditto Campaign on @zerufinance 's Zaps. You do not need one Giant Position to show up. The feed shows the same account stacking small BTC,SOL and ETH trades on Hyperliquid, and that adds up to about $2,830 in volume so far. The interesting part is how wide the payout list already is. All 48 traders on the live board have a prize showing next to their name right now, from $108.54 at the top to $1.09 at the bottom,and it grows as more volume comes in. Ditto closes Oct 29 at 16:00 UTC, so there are about four days left. Small, steady trades still have room to climb. app.zaps.wtf/campaigns/ditto

Yesterday's accepted list on the @vangrid_io bounty board reads like a little Road Trip a carved wooden chest in a nantes antique shop, $200. a walkin cooler wall in a cincinnati convenience store, $600. a Steel mezzanine inside a warehouse, $500. a raleigh campus dining hall filmed empty after hours,$1,000. a plant fence line and loading docks on madison's east side, $1,600. a 120 m street frontage, $600. a tampa city hall entrance, $1,500. a kerbless shared square in eindhoven, $1,705 eight bounties accepted on oct 23 utc, $7,705 in USDC between Them. across the whole board it's 969 of 1,003 bounties accepted so far what i like is how ordinary these places are. a robot learning to work indoors needs the cooler aisle,the dock gate, the stair up to the mezzanine. someone nearby with a phone can be that capture. three are still open, the biggest a $2000 station concourse to platform edge brief.

I really like how the GGC rewards discipline rather than Lucky Chance. The live Leaderboard on @gmgnai tracks realized PnL. According to the terms any open positions are fully factored into your total when the competition ends. therefore locking in your profits before the cutoff is the move. The @base track keeps the playing field level with a $2,000 cap and no top-ups allowed. No one can use an amount of money to rescue a bad trade. The final Rankings come entirely from your Entries, your exits and how calmly you manage your Position Sizing. Jesse Pollak has spoken about bringing Traders and Communities to Base. A competition, like this gives people a reason to step up and sharpen their Process. In the end the best exit strategy wins the Crown.

Say you film the back Stock room of a pharmacy with location switched off, the clip still goes through. I went back to the capture guide on @vangrid_io for this one. a Capture without location still works, it just never shows up on the map. some bounties also need location to check you filmed the right place, and those are the ones you'd Miss. there are two Locations in play. one is where the phone was when you filmed. the other is the network you upload from. the docs say uploads are Blocked in some Regions and that check looks at both. at ingest the network also checks that the Camera and Location signals agree. what gets anchored on Base is a content hash, a coarse geo hash and a Timestamp, with no Footage in it. so before walking a brief, turn location on and make sure the upload works from where you're standing A clip with no place attached is just a nice video.

A launchpad still collects its fee on a token nobody trades a Week later near the end of the new launch study from @zerufinance there's a fee line I barely saw anyone quote. pump (dot) fun's own docs list a 1.25% total fee on bonding-curve trades, 0.95% of it to the orotocol. that fee lands on every bonding-curve trade, whether the token lasts a month or An Hour. Now look at how long the trading lasts. across 399 Clanker launches in the same study, 7.5% had at least one transfer on day one. By day 30 it was 0.8%. The interesting part is how the activity markets design on zeruai.org moves who gets paid. real chain activity decides who mines, and miners earn when the market buys. No free inventory to dump, no team, no VC and no free miner allocation, and unsold Supply Burns. it is not live yet, registration is coming on Zaps. ask any launch one question, who still gets paid when nobody is buying?

you film a stairwell for a friend and the railing always hides one wall while the landing behind you never enters the Frame. that missing side becomes a hole in the model once the clip is processed on @vangrid_io . the app turns your walk-around video into a 3d model of the space. a capture that skips any reachable angle leaves a gap, so the guide calls for a slow steady pass that covers every side without quick pans. when you stop the clip uploads for an automatic quality check. clips That are too short or Unreadable get Rejected with a reason from the app and points only arrive after the Capture processes successfully. a rushed climb mostly leads to a wait followed by rejection. next time i would film the Landing on the way up and again on the way down so the turn Stays complete. does a quick phone video ever Leave you with an incomplete view of a Tight space?

A wallet that sent money to a sanctioned address is evidence. a wallet that sent money to that wallet is closer to a guess. The zScreen write up from @zerufinance treats those two very differently. a wallet one step from an address zScreen identified on its own can be raised to Review. It can never be flagged, at any sum on any chain, however often it repeats receiving alone never gets written down either. an address that deals with a huge number of counterparties, like an exchange or a bridge, is recorded as a boundary where the inference stops. In one replay,121 addresses ended up as boundaries. the interesting part is why they are this strict. one wrong call in the first generation becomes a hundred in the second, and those look exactly as confident as the real ones any address can be screened at zscreen.zeruai.org a guess that can only ask for review never gets to snowball into a verdict.

A bounty can reject your recording without making the effort useless. with @vangrid_io the requester picks a preferred submission and the selected contributor receives the Usdc. everyone else keeps ownership of their footage, which leaves room for that recording to be useful elsewhere. before filming starts, the reward is held on the Platform,letting you confirm the funds are available. submitting is free, and reconstruction costs are covered by the bounty. You can record a new walk or submit footage you already have. after A Submission wins, the requester receives a 3d model, while the person who filmed it retains the original clip. I like that the process makes ownership clear, even when a Submission loses. if your recording is not selected, it could still suit another nearby request.

wallet counts can make a campaign look busier than the trading really is on Aspen, the Zaps board from @zerufinance separates registrations from actual trading activity. each wallet needs its own volume, while wallets funded from one source are counted together. registrations may look impressive, but eligibility depends on trading volume. nearly 2,000 registered traders 5 accounts have traded about $4,281 in total volume $10,000 required per trader. the full $50K pool requires $15M in traded volume, and only about $14 is currently available. the leading account remains roughly $6,670 below the minimum. i like that the qualification math is public, although current activity is still limited. the interesting test is whether traders can meet the threshold through genuine activity.

you order groceries and the delivery bot turns around halfway because the curb it expected is now a construction barrier. that lag between what a machine remembers and what is actually there limits how useful its sensors and software can be. @vangrid_io is building a distributed network that collects and verifies spatial data so systems stay tied to conditions on the ground. it pulls fresh measurements from many points instead of leaning on one fixed dataset from months earlier. warehouses get rearranged. roads pick up new obstacles. buildings shift long after the first scan. a robot can still choose poorly if the picture it uses is already stale. how often does the map in your pocket already lag behind the street you are standing on? Image: phone screenshot of a navigation app showing a road segment marked with a recent construction barrier overlay.

Wall Street may have taken the weekend off, but Oracle, AMD, and Nebius were still getting attention on Hyperliquid. Ashutosh from @zerufinance pointed out that people were trading tokenized versions of these stocks over the weekend, earning Zaps and moving up the $50K campaign leaderboard. The interesting part is how Zaps works. It tracks verified onchain trading activity, not specific tickers. So whether you're trading $PENGU or AMD, eligible volume can count toward your score. Wash trading gets filtered out, so fake volume won't help. This creates a pretty different dynamic from traditional markets. Stock exchanges have fixed trading hours, but on Hyperliquid, the weekend can be another opportunity to earn points. The Ditto competition tracking these trades runs until October 29. Already trading on Hyperliquid? Check the leaderboard here: app.zaps.wtf/campaigns/ditto Traditional markets closed on Friday. The competition didn't.

If your starting balance decides your rank, your starting balance becomes the thing people fake. That's the bit of a recent research post from @zerufinance that stuck with me. Park a big deposit, trade for three days, pull it out, and a naive average makes it look like you started with a tenth of what you had. Your percent return suddenly looks heroic. The obvious fix is a minimum account size. They passed on it, because it deletes about a quarter of everyone who trades and draws a bright line for people to sit just above. The interesting part is what they used instead. It can only push a rating down, never up, and it touches roughly 2% of accounts. Small accounts can still come out on top. Seven wallets under $10K beat the typical million dollar earner. Worth connecting the wallet you actually trade from at app.zaps.wtf. Win by trading, not by dividing by a small number.

1.28M captures and only 5,100 attestations reads like almost nothing on the grid got anchored. That's not what the number means. On the explorer for @vangrid_io, an attestation is one Merkle root written to Base, and each root carries a whole batch of fingerprints. Divide it out and it's roughly 250 captures per root. The latest one, tree #5281, held 143 captures and 133 logins this morning. Any single capture can still be checked on its own. You take its fingerprint plus a short list of sibling hashes, rebuild the root, and compare it to the one on chain. Vangrid doesn't need to answer for that check to work. The interesting part is the trade they chose. Batching keeps anchoring cheap at this volume, but the chain only proves a capture existed by the end of its window, not to the exact second. Count what's inside the roots, not the roots.

Busy streets are not the problem. Every buyer already wants those filmed, so the bounties write themselves. The harder places sit inside the design of @vangrid_io. According to the docs, $VAN is meant to act as collateral and the settlement asset for the network, and the token is not live yet. The planned loop has three parts: - requesters may pay the platform fee in $VAN at a lower rate than by card - the collected fees fund bounties posted by Vangrid itself - those bounties target areas where the network still has no coverage The interesting part is the direction of the flow. Money paid by the demand side is not sent to holders. It is turned into incentives for the supply side, aimed at the gaps instead of the hotspots. A market fills the places buyers already know. A fee loop can pay for the places nobody has asked about yet.

A bot detector that flags a fixed share of every batch is not counting bots. It's grading on a curve. That's the argument in the latest @zerufinance research post. In one campaign cohort of 1,981 wallets, a relative-rank detector flagged 618 as hard bots, about 31%. The median flagged wallet had 1,886 transactions. The "bots" were the most active part of the group. A hand check of a sample found three of five genuinely automated. The other two were a power user and an airdrop farmer, both human. The interesting part is the fix. zScore scores a wallet from its own onchain history against a fixed reference, so the same wallet gets the same score in a cohort of 2,000 or 2 million. Penalties are graduated, so a false positive costs part of a payout, not all of it. Next time someone posts what share of an airdrop was bots, ask what that detector says about a group nobody suspects.