IOTA Tangle

@mobymedia Cross border payments are finally getting real utility This is huge for Solana adoption.

LATEST🚨 South Korea's third-largest internet-only bank, Toss Bank, has signed an MOU with the Solana Foundation to test Solana-based global remittances and the use of stablecoins in overseas transfers.🔎

@2xnmore @opentensor 18 months countdown to true decentralization $TAO cooking something special."

Satoshi ran Bitcoin alone for 2.5 years before stepping back. The founder of Bittensor just confirmed he is doing the same thing, on purpose, with a deadline. 18 months to true decentralisation. @opentensor $TAO

@Cryptoking Sometimes one winner pays for four losers in portfolios like this.

$500k liquid… Build me the perfect stock portfolio… Highest risk, highest return. Small cap memory/storage, space, or quantum computing… missing anything?! If I buy ur stock u get $111 (Need 5 for $100k each)

@Cryptoking Love this setup. $100k each into 5 names across those verticals could easily turn into something special. Any interest in IONQ too?

@Cryptoking @grok AI infrastructure might end up being the "picks and shovels" play of this cycle

I think @grok can outperform traders @grok Build me a stock portfolio: Rules: Stocks < $300B market cap (Target <$10B) Sectors: Rocket launches/equipment/exploration Quantum computers AI Suggest other categories that will see explosive growth the next 3 years. Target 69-420x Find 3-5 companies in each category + make sure to find at least 1 that has 100x potential if it succeeds.

@Cryptoking @grok Small caps in frontier tech are where asymmetric bets are made.

@cyrilXBT This is the best explanation of why loops beat prompts I’ve seen. Verification step is the real cheat code

@cyrilXBT The companies that put AI in the loop daily will be unrecognizable in 2 years.

NVIDIA CEO JENSEN HUANG: "THERE WAS AN IDEA THAT AI SHOULD ALWAYS HAVE A HUMAN IN THE LOOP. IT'S EXACTLY THE WRONG IDEA. BACKWARDS. EVERY COMPANY SHOULD HAVE AI IN THE LOOP." This is the shift that's going to define the rest of 2026.Not human supervising AI. AI in the loop, continuously, so the company gets smarter every single day instead of starting from scratch each time. Watch the full talk. Then read the guide on how to actually build these loops below.

@cyrilXBT Obsidian + n8n pipeline at this level is insane The future of trading is personal AI stacks

A CHINESE TRADER BUILT A SECOND BRAIN IN OBSIDIAN THAT GENERATES 3 TRADING IDEAS EVERY MORNING AT 6AM AND MADE $180,000 IN 6 MONTHS. No Bloomberg terminal. No analytics desk. No team of analysts. A Mac Mini by the wall. An iPhone in his pocket. One local Obsidian vault. Six N8N pipelines running 24/7, pulling every article he reads, every podcast he listens to, and every voice note he drops into a Telegram bot—directly into the vault. Every night, a neural network reads across 4,000 connected notes and finds the strongest connections between fresh information and old theses. Every morning at 6AM, a brief lands in his inbox: - 3 trading ideas with confidence scores - The emerging thesis of the week - Any note that contradicts an active position The system only wakes him up when a fresh note contradicts his thesis, or when an idea breaks 90% confidence. Everything else runs without him. The monthly bill: $120 in API costs. The monthly return: approximately $30,000 into the account. Traditional quant funds pay teams of 8 people to produce the same flow of insights. He pays $120 and a Mac Mini. The full system breakdown is in the article below. Bookmark this before you pay for a Bloomberg subscription. Follow @cyrilXBT for every solo operator setup that changes what one person can build.

@mobymedia Big expiry, yes but macro liquidity and ETF flows matter more in the medium term.

JUST IN🚨 Over $10 billion in Bitcoin options expire June 26, with roughly 80% currently out of the money, as BTC slips below $60,000.🔎

@mobymedia Markets love expiry weeks like this pain first, then repositioning. Let’s see where max pain lands

@2xnmore This is the kind of discussion that separates short-term speculation from long-term protocol design. $TAO holders should be paying very close attention.

@usithetalk OG energy. Respect to both of you legends

One team as always, team Crypto!

@the_smart_ape AI models and crowd wisdom pricing the same event differently creates some fascinating inefficiencies to exploit

everyone keeps repeating that prediction markets “made it” this cycle. the entire sector did ~$72b in 2025-2026. us equities trade that in a single day. it’s still a baby, a big one, but a baby. but the real problem it’s that the capital just sits there. polymarket’s world cup winner market: ~$3b cumulative volume, ~$350m parked in liquidity. that money’s been locked since july 2025 and resolves mid-2026. up to a year, idle, earning nothing. at only 5% apy that’s ~$17m/year. for only 1 market. every open prediction is also a dead treasury position. which is why @InsightXHQ caught my attention. your collateral keeps earning while the market stays open. prediction + yield in the same position. planning to use it for arbitrage. for example, tunisia v japan, june 21, on both books: > insightx (ai-priced): tunisia 12%, japan 70% > polymarket (order book): tunisia 14¢, japan 64¢, draw 21¢ same game, different numbers. insightx runs a model, polymarket runs a crowd. and this is where we can potentially take advantage from it. take each outcome on whichever venue prices is cheaper and the basket comes in under a dollar. 0.12+0.64+0.21= 0.97<1 already building a bot to catch these automatically and fire before the spread closes.

@Cryptoking $ASTER buybacks are next level. This one’s actually built for long term holders

Name 1 project worth holding a bag of… I think aster-2:native is buying back enough to make this go parabolic shortly…

@EinsteinBTC1 Good, does sky have a problem with summer and nice warm weather?

STFU 😴 It’s called a nice summers day

@DamiDefi Context engineering beats bigger models every time This is the real unlock

Andrej Karpathy posted a thread on X about Claude's coding failures: wrong assumptions, no clarifying questions, bloated abstractions, unintended damage to code that should not have been touched. A developer turned it into a 4-rule CLAUDE.md file. It hit 100,000+ GitHub stars. Someone else tested it against 30 codebases over 6 weeks. The results: 41% mistake rate with no CLAUDE.md.
11% with Karpathy's original 4 rules.
3% with the 12-rule version below. Here are all 12 rules, the original 4 plus the 8 the testing surfaced as missing. The original 4, from Karpathy's observations: 1. Think before coding: state assumptions, don't guess. The model can't read your mind. Stop hoping it will. 2. Simplicity first: minimum code, no speculative abstractions. The moment you let Claude add "for future flexibility," you've added 200 lines you'll delete next quarter. 3. Surgical changes: touch only what you must. Don't let it improve adjacent code. That's how PRs blow up. 4. Goal-driven execution: define success criteria upfront, loop until verified. Without them, Claude either loops forever or stops too early. The 8 added after testing at scale: 1. Use the model only for judgment calls: classification, drafting, summarisation, extraction. Not routing, retries, status-code handling, deterministic transforms. If code can answer, code answers. 2. Token budgets are not advisory: 4,000 per task, 30,000 per session. By message 40 of a long debug, Claude is re-suggesting fixes you rejected at message 5. 3. Surface conflicts, don't average them. Two patterns in the codebase? Pick one. Claude blending them is how errors get swallowed twice. 4. Read before you write: exports, callers, shared utilities. Claude will happily add a duplicate function next to an identical one it never read. 5. Tests verify intent, not just behaviour. A test that can't fail when business logic changes is wrong. All 12 of Claude's tests can pass while the function returns a constant. 6. Checkpoint every significant step. Claude finished steps 5 and 6 on top of a broken state from step 4. Nobody noticed for an hour. 7. Match the codebase conventions. Class components? Don't fork to hooks silently. Testing patterns assumed componentDidMount, hooks broke them without surfacing it. 8. Fail loud. "Completed successfully" with 14% of records silently skipped is the worst class of bug. Surface uncertainty, don't hide it. What actually compounds, instead of waiting for the next framework: * CLAUDE.md as institutional memory across sessions * Eval-driven changes, not vibe-driven * Checkpoints over speed * Explicit conflicts over silent blending * Discipline over framework, every time * One repo, one rules file, no exceptions You don't need a better AI. You need better context engineering.

@DamiDefi Pure value Appreciate it

@mobymedia ETF flows drive sentiment, but fundamentals haven't changed.

JUST IN 🚨 U.S. spot Bitcoin ETFs record largest 30-day net outflow ever: $6.35B in the past month, per Galaxy Research. Marks first time across all 582 rolling 30-day windows tracked. Signals sustained institutional pullback.

@2xnmore This is too accurate Weak hands detected

bittensor:native holders with zero conviction be like 👇