Mircea Dima 🦄

I recently had an interesting discussion, which was featured in the Financial Times with regards to how stablecoins are entering the financial mainstream - from Cross-Border Payments to Payroll. At MANSA we have built our business with #stabelcoins, as we see it as the most efficient way to transfer value real-time and to create capital efficiency across payments infrastructures and settlement layers. 𝗙𝗲𝗮𝘁𝘂𝗿𝗲𝗱 𝗵𝗲𝗿𝗲: https://on.ft.com/4e3oq6r

37

4 Comentarii

Jeff Bezos is raising $100B to acquire manufacturing companies and automate them with AI. Here's a breakdown of his genius M&A plan👇 Recently, Bezos founded Project Prometheus. The startup builds AI systems that learn from real‑world factories, machines, and experiments to make manufacturing efficient. Since its launch in November 2025, it has already been funded with about $6.2B. Now, Bezos is in early talks to raise $100B, actively traveling and fundraising globally. The plan is simple: → Raise a mega‑fund → Buy manufacturing companies → Upgrade them with an AI/automation platform. The focus is on struggling manufacturers in sectors such as chipmaking, aerospace, and defense. This is textbook vertical integration. • own the capital (the fund) • own the physical layer (factories) • own the intelligence layer (Prometheus) Bezos is building a system where capital, software, and operations are tightly integrated. For M&A professionals watching, there are a few clear implications: 1) A new mega‑buyer $100B dedicated to old, under‑automated manufacturers makes Bezos a deep-pocketed bidder. He can easily outcompete traditional PE firms and strategics in this space. 2) Increased value for manufacturing companies Valuations for industrial assets will go up. Especially for those that are AI-friendly with large process data and repetitive workflows. Companies that sold for 7–9x EBITDA will see premium pricing if they fit an AI-upgrade thesis. This signals a serious push to build an AI-powered industrial giant. If it works, this could be one of the boldest M&A bets in the AI era.

17

Most DeFi interfaces are too slow for the next 100M users. So, I built a solution. ⚡️ I’m thrilled to announce that I’ve officially shipped InstaInjective for the Injective Africa Buildathon! The Problem: New traders on Injective often struggle with complex UIs, multiple confirmation clicks, and the friction of advanced order types. In a market where every second counts, high-latency UX is a dealbreaker. The Solution: InstaInjective is a 1-click trading terminal on Injective inEVM. 1.Minimalist UI: Abstracts away contract complexity. 2.Seamless Execution: Users connect and trade with a single confirmation. 3.Non-Custodial: Power and speed without sacrificing decentralization. Building on Injective’s cutting-edge infrastructure has been an incredible journey in scaling DeFi accessibility. 🙌 Shoutout to my amazing team who made this possible: 👨‍💻 Devendra Yadav 👨‍💻Prem Borde 👩‍💻 Rashi Gupta Each one of you brought incredible energy and skill to this build. This ship belongs to all of us. 🚀 Check it out here: https://lnkd.in/g-CpVRWc Github Repo: https://lnkd.in/g6s9kYxW Special thanks to @Injective and @DoraHacks for the platform to ship and learn! #Injective #Web3 #DeFi #Blockchain #Solidity #InEVM #Buildathon #DoraHacks #CryptoEngineering

26

2 Comentarii

🚨 In 2026, the biggest winners in DeFi won’t be protocols. They’ll be developers who understand Cross-Chain architecture. Here’s why. In 2025 alone: • $11.4T+ in DEX trading volume • $2.8B+ lost in bridge exploits • 50+ chains competing for liquidity Yet DeFi still operates like isolated islands. Users today split capital across multiple ecosystems: • Security on Ethereum • Speed on Arbitrum • Low fees on Polygon This creates fragmented liquidity, poor UX, and massive security risks. And bridges? They've become the largest attack surface in DeFi. ⚙️ The real shift happening right now The industry is moving from token bridges → cross-chain messaging infrastructure. Protocols like Chainlink’s Cross‑Chain Interoperability Protocol (CCIP) allow smart contracts to: • Send verified messages across chains • Execute logic remotely • Synchronize protocol state between networks This changes how we design DeFi systems. 🧠 While researching this, I explored a real architecture challenge: Building a Cross-Chain Rebase Token The problem is subtle but dangerous. If the same token runs on multiple chains, supply can diverge. Example: Ethereum supply → 1,000,000 Arbitrum supply → 1,000,050 Polygon supply → 999,980 Small timing differences create arbitrage opportunities that break token economics. 💡 The solution is a distributed systems pattern: Master–Slave Architecture • Ethereum → Source of truth • L2 chains → Execution layers The master contract calculates the rebase and sends CCIP messages to synchronize supply across networks. Result: ✔ Deterministic supply across chains ✔ No liquidity fragmentation ✔ Consistent token economics 📊 The biggest realization for me: The next generation of Web3 engineers won’t just write smart contracts. They’ll design distributed financial systems. Meaning developers must understand: • Asynchronous architecture • Cross-chain message flows • Finality differences between chains • Multi-network testing environments The real question for 2026 is no longer: ❓ Which blockchain will win? But: 💡 How do we architect protocols that use multiple chains as one system? I summarized the full research in the carousel below. Curious to hear your perspective: Will DeFi evolve around a liquidity center like Ethereum… or become a federation of specialized chains connected together? 👇 Let’s discuss. #DeFi #CrossChain #CCIP #Chainlink #SmartContracts #Solidity #Web3 #Blockchain

5

1 Comentariu

Excellent article from Tomasz Tunguz about the future of the token economy "Technology companies are adding a fourth component to engineering compensation : salary, bonus, options, & inference costs. Levels.fyi pegs the 75th percentile software engineer salary at $375k. Add $100k in inference & the fully loaded cost is $475k. That’s 21% in tokens. The question CFOs will pose : what am I getting for all this inference spend? Can I do it cheaper?" It is the core reason why we built Edgee AI gateway : Cut and control your cost: Cost dashboard in real time, token compression, LLM selector. https://lnkd.in/dXmqVabR

6

From Neobank to Institution: Revolut got a UK banking license + £3B investment commitment. Crypto-native fintech → regulated bank. That trajectory is now proven. Not theoretical Nyla (Shari'ah-compliant neobank) launching Ghana next quarter. Islamic finance embedded in neobank UX. Ualá adding stock trading to Mexican neobank. Feature expansion signal: neobanks are becoming wealth platforms Path is clear: simple → compliance → scale → institutional products → banking license. Regulators now see a playbook. Emerging-market neobanks have a template The edge for LATAM operators: regulatory expertise + local KYC + stablecoin rails + Simple products runing on a efficient and lean infrastructure. That combination unlocks regional dominance. Next wave: emerging-market neobanks as primary financial infrastructure

2

Allium - Stablecoins: The emergence of a new payment rail 🔹 Payments Are Entering Their Growth Phase 🔹 The Domestic Thesis Is Replacing the Remittance Narrative 🔹 Velocity Signals Infrastructure Maturity 🔹 Holder Distribution Reflects Functional Adoption 🔹 Multi-Chain Specialization Is Structural, Not Fragmented

18

Y Combinator Launches of the Week: (This round-up is for informational purposes only, and the order is based on the date/time of the launch, from earliest to latest) 1) Sciloop (YC F25): Founded by Bilal Asmatullah and Osman Siddique, Sciloop automates research, turning seed ideas into publication-ready papers. Know More At: sciloop.dev 2) Aspect (YC F25): Founded by Gurish Sharma, Bright Xu, and Edison Chiu, Aspect is a platform with AI agents that understand a team’s video library and workflows better than humans. Know More At: aspect.inc 3) Rivet: Founded by Sam Gorman, Rivet helps product designers change software without coding. Know More At: rivet.design 4) Serra: Founded by Alan Wang and Albert Stanley, Serra is an AI recruiter that fully automates candidate sourcing and outreach, finding ideal talent from LinkedIn, GitHub, Crunchbase, and your ATS. Know More At: serra.io 5) Alloy: Founded by Simon Kubica and Christian Iacullo, Alloy is AI Prototyping for Product Management. Know More At: alloy.app 6) Dome (YC F25): Founded by Kurush Dubash and Kunal Roy, Dome is a unified API for prediction markets, like Kalshi and Polymarket. Know More At: domeapi.io Post: https://lnkd.in/ggvfCH5c X: https://lnkd.in/gs34UtR7 Source: https://lnkd.in/gBBjGht8

5