[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]

@px_721 Avatar @px_721 px

px posts on X about $zksl, $reppo, ai, preference the most. They currently have XXXXX followers and XX posts still getting attention that total XXXXX engagements in the last XX hours.

Engagements: XXXXX #

Engagements Line Chart

Mentions: XX #

Mentions Line Chart

Followers: XXXXX #

Followers Line Chart

CreatorRank: XXXXXXX #

CreatorRank Line Chart

Social Influence

Social category influence cryptocurrencies #5362 finance XXXX% stocks XXXX% automotive brands XXXX% social networks XXXX%

Social topic influence $zksl #11, $reppo #4, ai 8.62%, preference #100, future 5.17%, solana 5.17%, spam #661, prediction #1904, ceo #2954, stack #1358

Top accounts mentioned or mentioned by @repponetwork @reppo @rgvrmdya @px721 @m4cero @bhaskarchain @cobreonchain @hydefella @newlab @lurkaroundfind @scobleizer @blevlabs @levangielabs @0xpx10whosaroundreppobcf56912b49a @0xpx10industrialisingtherarestcommodityinai4c9f8579b989 @orixcrypto @turtleonchain @ctrlpmoney_defi @ricksuperprick @xbrazilzz

Top assets mentioned zkSealevel (ZKSL) REPPO (REPPO) Solana (SOL) Developer Camp (DEVELOPER) Camden Property Trust (CPT) machines-cash (MACHINES) StarkNet (STRK) Tesla, Inc. (TSLA) Bittensor (TAO) Chainlink (LINK) Celestia (TIA) ZetaChain (ZETA)

Top Social Posts

Top posts by engagements in the last XX hours

"missed this pack it up claimed true: real stark proofs a trace permutation constraints state bindings writable lookups state-root proofs verifiable progress NONE of these appear in repo not even partially $zksl"
X Link 2025-11-23T16:19Z 2381 followers, 3248 engagements

"$cpt in the right rooms newlab = serious well funded deep tech venture platform at the intersection of mobility drones & energy newlab + new system ventures deploy capital into member startups membership = - access to prototyping tools and engineers - lab / office space (brooklyn or detroit) - low friction way to get a physical US base would be a good US launchpad enabler detroit location has unique aerial pilot space and is actively showcasing drone tech from the newlab roof - did a drone donut delivery demo recently could empulser here: - install early hive pads - run autonomous drone"
X Link 2025-12-04T00:24Z 2381 followers, 4557 engagements

"above would place reppo at the base of future ai stack ai stack = X layers compute models agents memory training data applications reppo is placing itself underneath all of them. training data is the only layer that touches all other layers and which every model and every agent depends on everyone else has internal pipelines - not protocols not markets not verifiable systems not open contributor ecosystems"
X Link 2025-12-08T21:19Z 2376 followers, XXX engagements

"to me reppo are saying agents require continuous data to update = that data must have provenance / incentives / structure / liquidity - we are the coordination layer that makes this possible filecoin = indispensable for storage if successful reppo = indispensable but for something far more valuable"
X Link 2025-12-08T21:20Z 2378 followers, XXX engagements

"Claim X - stark proved XXX% complete = FALSE Claim X - on chain program XXX% deployed = half true but meaningless Claim X - orchestrator & indexer XXX% operational = partially true but not the real work (easiest part of stack to fake if faked) Claim X - CI/CD XXX% automated = true but irrelevant - this is not a technical milestone Claim X - simplified AIR XX% coverage = FALSE - zero public evidence - not even the XX% exists core part of STARK system - nentioning a trace permutations + lookups is narrative gloss not implementation Claim X - documentation & testing XXX% comprehensive = FALSE -"
X Link 2025-11-23T13:30Z 2376 followers, 4616 engagements

"$zksl both fake statements pushed by official team quote is from most recent update we are representing serious techs"
X Link 2025-11-23T15:59Z 2376 followers, 3030 engagements

"tldr for my post that is about to come on this baby food $zksl where are the constraints - your whole wall of text avoids the only thing that matters - no air no constraints no correctness"
X Link 2025-11-23T19:02Z 2376 followers, 1455 engagements

"let me eat this baby food $zksl (this is my final post - bored) you (ai) is describing the pipeline not the proof wiring rpc - json - winterfell - on chain metadata is easy proving sol state transitions is the part missing this instantly shows you (ai) dosent understand the difference winterfell proofs mean nothing without air - its compression not verification = you are celebrating zero cryptographic guarantees you (ai) cited stark verify as if it proves correctness - it dosent - it checks internal constitency of empty circuit - dosent verify sol = whole argument collapsed but will continue"
X Link 2025-11-23T19:04Z 2376 followers, 3472 engagements

"seeing files isnt seeing proofs a pipeline isnt a prover - and thats fine the issue is the official team is presenting it as if it is $zksl"
X Link 2025-11-23T19:08Z 2376 followers, 2970 engagements

"$zksl if u cant see a disconnect u r disconnected from source"
X Link 2025-11-23T21:18Z 2376 followers, 3826 engagements

"anyone have input re below / $machines X words have me exit for now (I like everything else)"
X Link 2025-11-29T15:49Z 2376 followers, 1730 engagements

"$reppo - @reppo recent learnings X - another new category claim alongside compute bottlenecks storage exists compute exists but nothing coordinates the production verification and monetisation of data that trains models = there is no protocol layer for training data literally nothing else is attempting to build this training data protocol reppo is solving a problem no one else is addressing the economic coordination of training data"
X Link 2025-12-08T21:18Z 2378 followers, 4965 engagements

"kled: simple do tasks get paid - scales by adding workers. heavy ops + QA. labour driven reppo: signal preference. scales via staking + consensus not headcount - reusable across domains - human + agent co training over time (see thread)"
X Link 2025-12-09T16:07Z 2379 followers, XXX engagements

"signals ( $reppo ) - chose commit reveal because fhe/zk privacy infra isnt production ready - evaluated miden starknet other chains - but base has the users and activity they actually need -zama may unlock encrypted voting and realtime privacy in the future - reppo plans to be an early adopter signals to me - practical technical decisions (not chasing - i like)"
X Link 2025-12-01T09:34Z 2372 followers, 3064 engagements

"$reppo vs tao (bittensor) tao subnets collapse when emissions drop - reppo does not tao lacks privacy - reppo will integrate fhe when ready tao cannot handle enterprise - reppo private subnets will tao rewards compute - reppo rewards correctness tao doesnt produce reasoning - reppo does"
X Link 2025-12-02T15:39Z 2374 followers, 2408 engagements

"my main questions re $reppo as of now are below I have confirmation from the team they have been passed along to relevant individual today I have no worries either @Repponetwork"
X Link 2025-12-02T16:01Z 2374 followers, 2164 engagements

"$reppo - @reppo X. llm helper for votes / pod boosting bribes = right direction if they want millions of micro judgements without burning users out / good for revenue and engagement X. commit reveal ux option X with opt in per user = best maximises label quality / future flexibility whilst keeping suspense in front end near side thinking = - voter UX (llm helper) - monetisation (bribes / boosting) - ux / quality tension (commit reveal debate)"
X Link 2025-12-08T21:29Z 2372 followers, 2170 engagements

"graywhale = one of first companies to use impending auto buy feature to rebuy $developer with profits"
X Link 2025-12-02T13:39Z 2372 followers, XXX engagements

"@reppo anti spam - subjective adversarial high financial risk adult content - preference heavy culturally divergent rlhf dominant robotics - edge case heavy continuous feedback domain voice ai - subtle tonal preference sensitive"
X Link 2025-12-06T15:45Z 2376 followers, XXX engagements

"forgot to add section of my research: CMS: conviction is huge (front run narrative peaks) very early to solana chainlink (oracles) avalanche helium + lots more MH: XAI (elons L2) io net celestia eco zetachain CV VC: seba bank (now equity bank - swiss crypto bank) XX shares early to solana also + near Protocol: IPFS filecoin ceramic / compose db"
X Link 2025-12-07T20:01Z 2372 followers, 2501 engagements

"$reppo @Repponetwork confirmed current verticals: - robotics (manipulation embodied agents failure data rl/dpo datasets) - adult entertainment (private subnet for content preference data) - coding (vibe coding platform sourcing training labels) - market research (enterprise labelling + preference data) - geospatial (early subnet alignment) - email security / spam detection (legit private subnet for bec + ai spam classification)"
X Link 2025-12-05T23:08Z 2381 followers, 4030 engagements

"$reppo @reppo judgment heavy verticals where human disagreement is high traditional datasets are fragile / expensive / bot corrupted again - good verticals where prediction market truth inference has advantages"
X Link 2025-12-06T15:41Z 2381 followers, 2507 engagements

"who's around @reppo some research and thoughts on this slow sunday"
X Link 2025-12-07T19:06Z 2381 followers, 9107 engagements

"Industrialising the Rarest Commodity in AI @reppo"
X Link 2025-12-09T15:04Z 2381 followers, 21.1K engagements

"appreciate you taking the time to read / share your thoughts high barrier / niche true today = not structural prediction market ux is harder - current users skew technical front end can hide markets: pick better answer / swipe / spam check backend turns clicks into stake weighted labels ux already improving: cleaner commit reveal llm helpers pod boosts ux is a risk but mainly a product problem not a structural blocker"
X Link 2025-12-09T16:07Z 2381 followers, 1706 engagements

"$reppo / @reppo article from irys on reppo tightens previous thesis that theyre aiming for the coordination layer for training data further signs towards strong infrastructure confirmation - not hand waving about provenance / programmability storage cost / performance is unlikely to be a bottleneck for them now"
X Link 2025-12-10T21:15Z 2381 followers, 3636 engagements

"economies markets + data economy framing past on chain data markets didnt work because they were just places to upload files = not systems that create and maintain useful data static CSVs + token incentives doesn't = a functioning data supply chain failed because: - the data wasnt created in protocol - wasnt continuously refreshed - it wasnt immediately usable by models or agents - there were no native incentives to keep quality or cadence high further clarification on proposal above (something structurally different) - data is produced inside the protocol not uploaded after the fact -"
X Link 2025-12-11T20:18Z 2381 followers, XX engagements

"further @rgvrmdya framings compute = abundant scraped data = abundant synthetic data = abundant human preference & alignment signals = scarce & expensive prediction markets as core architecture: four mechanical advantages - reward accuracy not throughput - get confidence weighted signals - global participation - turn alignment into a tradable asset"
X Link 2025-12-11T20:18Z 2381 followers, XX engagements