DeFi Scholar 🎓🎓

@Fr0oZi @grvt_io yeah I did

@ModestusOkoye @grvt_io Wow, did you make this dashboard yourself?

Want a quick glance at what’s happening onchain for @grvt_io? Built a simple dashboard for exactly that. It tracks volume, TVL, and OI, while also comparing GRVT against other perps. More onchain metrics will be added over time, but this is already a very solid start. Check it out here: defillama.com/pro/ckcgd5lcdd…

@holly_web3 @StandX_Official avoid StandX. that's my humble advise

I applied for the @StandX_Official ambassador program on Kaito and got accepted. But honestly, the rewards don’t look that great. They’re offering $1,000 per month, but asking for something like $10M in trading volume. Come on, you’d end up paying more than $1,000 in fees anyway. This feels more like a fee discount for airdrop farmers rather than a real reward. I expected the team to be more generous. Anyway, hope they succeed at TGE. -> standx.com/referral?code=…

@TheDeFinvestor seems there is still memecoin demand

@ModestusOkoye This is really surprising tbh I didn't expect them to do so well during a bear market

Most interesting trends in DeFi rn: • 24/7 equity markets (HIP-3 coded) • Tokenization and RWA looping • New lending primitives (Aave V4, Silo V3, and soon a new Morpho product) • crypto neobanks gaining huge traction Anything else?

@LiquityProtocol Vesting cliff reached

BREAKING: Circle has acquired Liquity. This acquisition will enable Circle to offer its users a non-freezable stablecoin and directly distribute yield under the Clarity Act.

@Eli5defi what a day to be alive

It's a wild day for Anthropic and AI. Anthropic leaked its own source code. Twice. In one week. 512,000 lines of TypeScript. 1,900+ files. Now permanently public. Here's what happened ⤵ -- > THE TIMELINE → March 26 A server setup mistake exposed about 3,000 internal Anthropic files about an unreleased next-gen model. It was codenamed Mythos (internal nickname: Capybara). People archived the files at m1astra-mythos(dot)pages(dot)dev (v1 + v2 docs became public), though it may be taken down by now. → March 31 The npm package `@anthropic-ai/claude-code` v2.1.88 shipped with a huge 59.8 MB “source map” file. That file pointed straight to the full, readable TypeScript source code, which was sitting in a public Cloudflare R2 bucket. Chaofan noticed it first and posted about it on X. Within hours, the community made lots of copies fast. A mirror repo, instructkr/claw-code, reached 72.5k stars and 73k forks. It reportedly hit 50k stars in about 2 hours, one of the fastest ever. → April 1 “April mop” jokes spread because one leaked feature had a hardcoded rollout window of April 1–7 in the source. -- > What’s actually inside > Architecture (in plain English) ❶ The “brain wrapper” A 12-step control system (12-Harness) that tells the AI how to plan, check its work, and use tools safely. this is the main advantage of the Anthropic model (see @yq_acc post below) and I will talk about this later. ❷  Lots of planned built-in tools 40+ tools the AI can use, includes a safety checker that analyzes bash commands before running them (2,679 lines of code) ❸ Smarter memory for long tasks A 3-level system that compresses and summarizes context so the AI can keep working for hours without losing track > Unreleased features ❶ KAIROS A background helper that watches your code repos and can act proactively, without you asking ~ probably the endgame for current AI models. ❷ Dream “sleep mode” where it organizes and strengthens its memory while you’re away ❸ BUDDY A virtual pet game with 18 species and gacha-style rarity with planned launch: April 1–7 > Other internals - Undercover mode: lets Anthropic employees submit AI-generated code to public repos anonymously - A “frustration chart”: tracks profanity (via regex) and reports it to an internal dashboard - Anti-copy traps: decoy logic meant to confuse competitors trying to replicate the system -- So far, Anthropic says this was a packaging mistake, not a breach. If you’re curious, there are some of the best community resources for the forked code out there. 😉

@waleswoosh nah. Arbitrum cooked massively then

No good airdrop has ever come from a crypto project with a blue logo. Green: Hyperliquid, Kaito, LooksRare. Red: Optimism, Uniswap (pink, but it counts). Yellow: Blur, Bonk. But blue? None

Saw some people panicking or asking about quantum computing's impact on crypto. At a high level, all crypto has to do is to upgrade to Quantum-Resistant (Post-Quantum) Algorithms. So, no need to panic. 😂 In practice, there are some execution considerations. It's hard to organize upgrades in a decentralized world. There will likely be many debates on which algorithm(s) to use, resulting in some forks. And some dead project may not upgrade at all. Might be a good to cleanse out those projects anyway. New code may introduce other bugs or security issues in the short term. People who self custody will have to migrate their coins to new wallets. This brings to the question of Satoshi's bitcoins. If those coins move, then it means he/she is still around, which is interesting to know. If they don't move (in a certain period of time), it might be better to lock (or effectively burn) those addresses so that they don't go to the first hacker who cracks it. There is also the difficulty of identifying all his addresses, and not confuse with some old hodlers. Anyway, it's a different topic for later. Fundamentally: It's always easier to encrypt than decrypt. More computing power is always good. Crypto will stay, post quantum.

@Cointelegraph okay.. another volatility day

🇺🇸 JUST IN: Trump to address the nation on the Iran war at 9PM ET Wednesday per White House Press Secretary Leavitt.

@0xTakeProfits I am telling you bro

@ModestusOkoye Feels accurate tbh lol

this is what next 2 years will be like

Want to know who is buying or selling $MNT and when? I built a simple dashboard on @Dune that simplifies that. The dashboard is so interactive that you can check for: > top buyers or sellers > filter it by period (7D to 60D) > filter it by holder's retention (maybe you need diamond hands or partial sellers) > search a specific wallet. Something huge is coming soon for all EVM tokens!

@shortsqueeznews might have to buikd a dashboard tagged "who are still buying cards"

The Pokemon card index is up ~50% this year. Last month, Logan Paul sold a rare Pikachu card for $16M, the most expensive trading card ever sold at auction.

@ashen_one marketing vault haven't worked. infint tried it with prompt to DeFi

@ModestusOkoye it’s as easy as promoting tho and yea ofc you gotta market it

For everybody using AI to try and make Polymarket bots or trading bots, this is for you Although it's gotten a little easier to make those things, they're still not to the point where you can actually cook with them autonomously One of the best ways that you can plug in AI to crypto stuff is through the fact that they're MUCH smarter than us in research and strategies and not just actual execution For example, with RoboNet, they do vaults where they deploy capital across different ecosystems and earn yield on it It could be vaults for staking, trading, loans, arbitrage, literally anything You don't have to do any research on how you want to make it. You just have your agent create the vault and let other people use it The Money Plan: Get your agent to do hours of deep research into existing profitable Hyperliquid vaults, other vaults that have failed, and top-performing vaults for the past ten years. Create one banger vault that people can deposit capital into. Then, based on the vault's performance from whoever else deposits into it, you earn fees You REALLY don't need to be doing two-second micro-trades and arbitrage between PolyMarket or anything crazy like that. You just have your agent do deep research and make a vault that can give people 10% a year And if that vault ends up having like $1,000,000 in it, you make $$$

@stacy_muur @DonutAI DO

I've been testing @DonutAI for the past few days. It's a Telegram-based AI agent that analyzes and trades through 152 tools across research, perps, DeFi, onchain data, prediction markets, and more. The setup is straightforward: • Sign in with an invite code • Click on "Get Started on Telegram" • Deposit funds from Solana/Ethereum/BNB/Arbitrum • Use the agent to analyze/trade the market The first thing I did was ask the agent what I should trade today. It ran a macro scan, analyzed sentiment and liquidity conditions, checked institutional flows, and returned structured reasoning on why it's unsafe to trade at the moment. Really like what I'm seeing so far, so definitely keeping it on my toolkit. If you're interested in trying it out, just comment "D0" below and I'll give out a few invite codes.

@pankaj I read this post and I didn't see what happens with the $33 million raised. Not saying that workign on something that is obviously not working is cool but then accountability is really hurting our space.

1/ We’ve made the difficult decision to wind down yupp.ai. The website will be up for another 15 days during which time users can download their chat data. New users won’t be able to sign up and existing users won’t be able to create new conversations after today. Yupp is a loved product by many and we are sorry to the community for this outcome.

@waleswoosh this is great wale... in addition to that, if you can handle that, a weekly or monthly space for people looking to pitch there ideas will be great too

StartupCash is an aggregator that we will be updating weekly with new, real funding opportunities for builders in crypto, stay tuned

Getting your idea funded is easier than you think. To prove it, I've launched an AI that estimates the value of your crypto startup idea. Pitch your idea to me: startupcash.io/pitch

@stacy_muur we need to wake him up

Almost $350K has been sent to Satoshi's wallet by random people in the past 12 months. Someone sent him $181K in a single transaction two months ago. Why 😭😭😭

@Nick_Researcher @Mars_DeFi same as me. the solution will fly out

@Mars_DeFi QC is no threat imo

Turns out the clock is ticking on all of us and we need Post Quantum Cryptography (PQC) sooner than was anticipated in this article below. Earlier, I highlighted some good post quantum standards that could be widely adapted in the coming months. • ML-KEM ( Module-Lattice Key Encapsulation Mechanism ) • ML-DSA ( Module-Lattice Digital Signature Algorithm ) • SLH-DSA ( Stateless Hash-based Digital Signature Algorithm ) • STARKs Here's what you need to know about them — • ML-KEM (FIPS 203): This is a standard based on CRYSTALS-Kyber. Its role is simple but critical. It alllows two parties to securely agree on a shared secret over a public network. That secret is what encrypts everything that follows. Think of it as: “How two strangers safely create a private channel.” It’s lattice-based, relying on problems that are currently infeasible for both classical and quantum computers to solve efficiently. • ML-DSA It is based on CRYSTALS-Dilithium. This is the new standard for digital signatures. It answers two questions: Did this message really come from the sender? Has it been altered? This is what secures: • software updates • blockchain transactions • authentication systems Think of it as: “Proof that this came from me and hasn’t been tampered with.” Why are ML-KEM and ML-DSA called lattice-based? Because their security comes from solving very hard problems in structured high-dimensional math called lattices, specifically module-lattice problems. • SLH-DSA It is derived from SPHINCS+ which is also used for digital signatures but built very differently. Instead of lattice math, it relies on hash-based cryptography. This matters because it introduces diversity. If lattice-based systems ever face unexpected weaknesses, SLH-DSA acts as a fallback. Although it has some tradeoffs ranging from larger signatures to slower performance. But it is stronger from a risk management perspective Think of it as: “A different lock, built on completely different physics.” So in simple terms: ML-KEM secures communication ML-DSA verifies identity SLH-DSA is a backup verification system • STARKs These are proof systems used to verify computation efficiently, especially in scaling systems like Starknet and other zero-knowledge-style architectures. They are often described as post-quantum or quantum-resistant because they rely on hash-based techniques and avoid some of the structured assumptions used by other proof systems. STARKs may be quantum-resistant, but the overall system still depends on other components like signatures, key management, hashing, and protocol design. This is because quantum security is end-to-end, not just one layer. — Overall migrations like this take years when you consider how long it will take to go from Standards -> implementations -> audits -> adoption -> key rotation. We may be early but it has already started and it is good that we are replacing the existing cryptography frameworks with the cryptography of the quantum age because it’s necessary.