limfx8

@UniofOxford / PE and special sits / Crypto and DeFi since 2017

[Ryan Watkins @RyanWatkins_]

Think we are entering one of the healthiest market environments for crypto in a long time with a small number of quality assets leading the way.

For the first time in over a year market participants are no longer reflexively bidding memecoins and other speculative nonsense at the first signs of risk on.

Simultaneously, there is growing alignment around owning real projects again with dispersion increasing between projects w/ PMF + growth, and projects that are vaporware / scams.

I guess after getting rinsed 80% - 90% so many times, retail is finally learning their lesson.

While I still think the majority of the asset class will struggle given how high the bar for allocation is in a world where anyone could invest in alternatives like AI equities, I do think that whatever capital does flow into crypto from here will concentrate into a small number of quality assets that have the potential for multi-bagger outcomes.

TL;DR: It’s a good time to be an asset picker.

[Hyunjin Kim @hyunjinvkim]

🚨 Excited to share a new working paper! 🚨

AI can improve individual tasks. But when does it improve firm performance?

Our paper proposes one key friction firms face: the "mapping problem" -- discovering where and how AI creates value in a firm's production process.

🧵1/

[Martin Varsavsky @martinvars]

AI is big but people's bodies are bigger.

Two molecules vs the entire AI industry.
Semaglutide: ~$33B in 2025. Tirzepatide: ~$36B. Combined: nearly $70B in actual revenue from two drugs to lose weight.

OpenAI: $20B annualized run rate. Anthropic: ~$5-9B. Combined: under $30B, and those are run rates, not collected cash.
A single injectable drug generates more revenue than the most valuable startup in history. Novo Nordisk runs at 40% net margins. OpenAI burned $8.5B in cash last year to generate its $20B.

Biology still commands an economic reality that software has not reached. Healthcare is $4.5T in the US alone. When you solve a problem that affects hundreds of millions of human bodies, nothing else competes.
The smartest bet is at the intersection: AI applied to medicine. Like the work we do building fully autonomous MDs at @certumaai

[veloX @veloXxn]

In my opinion, the real turning point here isn’t whether vibecoding is good or bad, but the segmentation of software development into layers.

> Generating ideas is now the cheapest part > Building the first prototype is the easiest it’s ever been > But security, architecture, maintenance, and scaling are still expensive

That’s why we end up with a strange situation:

One person can build a working product over the weekend. But once users see that product, it stops being a hobby and becomes a responsibility.

Vibecoding is incredibly powerful right here: It democratizes speed. But it doesn’t automate quality.

Node.js prototypes, open-source repos, quick demos, the Expo vs. Flutter debate… All of these are valuable because they lower the entry barrier. What’s missing in Turkey isn’t just getting started, but the discipline to iteratively improve what’s already started.

I think the new game is this: With AI, everyone can build an MVP. The winners will be those who manage what comes next.

In other words, not just those who can write prompts; but those who understand at what point an engineer, designer, security expert, and product strategist are needed.

In short, vibecoding isn’t a bubble. But it’s not just product development knowledge on its own either. It opens the door today, but tomorrow, only those who can build systems to stay inside will succeed.

[hoeem @hooeem]

https://t.co/4tchgZa4oC

[kanav @kanavtwt]

the meme has been updated

[Anton Osika @antonosika]

Lovable just crossed $100M ARR in 8 months.

Faster than OpenAI, Cursor, Wiz, and every other software company in history.

Today we're launching a game-changing update, it reduces the error rates by 91%.

Introducing Lovable Agent:

[Aadit Sheth @aaditsh]

The most dangerous addiction today isn't a substance.

Research on 100,000 people confirms that heavy short-form video use is just voluntary cognitive decline. We are actively training our brains to fail at hard tasks.

If you can simply sit with a problem for 10 minutes without swiping, you have a massive competitive advantage.

Basically, boredom is the new IQ.

[Philipp Schmid @_philschmid]

we now have a @GoogleDeepMind Luma page for events where the team is going to https://t.co/jSnyucNWjZ

[limfx8 @limfx88]

@deepcryptodive @justinalick @GearboxProtocol define minimal risk for >50% apr on eth? ^_^

[Stani @StaniKulechov]

Most interesting part was that the Bitcoin collateral was real Bitcoin, no wrappers.

In 2026, we will bring ETHLend back, I promise.

[MacroScope @MacroScope17]

Very interesting report today about this BTC purchase. Bloomberg got a rare on-the-record statement from Abu Dhabi Investment Council about it:

We see Bitcoin playing an increasingly important role alongside gold,” the ADIC spokesperson said in response to Bloomberg News queries. “Both assets contribute to diversifying our portfolio, and we expect to hold them as part of our near and long-term strategy.”

https://t.co/ChRe3kPSuQ

[matthew sigel, recovering CFA @matthew_sigel]

Who needs tokens?

$WSHP WeShop +521% to $5B market cap as post-IPO squeeze ignites.

The UK social commerce platform turning shoppers into shareholders proves you don't need a token to programmatically distribute rewards via equity:

Earn WePoints for buys/referrals/reviews → points vest in trust → convert to real shares after ~395 days of engagement.

Loyalty coded in, no blockchain required.

[Coffee Capital @Coffee__Capital]

🇺🇸 CDS MARKET FLASHING WARNING SIGNS:

CDS spreads continue to surge for AI companies. Oracle, CoreWeave 10Y CDS spreads are up over 30% in the past month alone, with other major players also up double digits.

We are seeing a ton of heavy, upfront "capex" spending by these companies. Not only that, but these hyperscalers are tapping the credit market for this capital at a scale we haven't truly seen before. The ultimate return on these investments is becoming more uncertain, hence investors are bidding up protection against the credit.

$ORCL $CRWV $AMD $INTL $AMZN $MSFT $GOOG $IBM

[Strobe Ventures @strobefund]

Kicking off Strobe Summit with a friendly game of padel, gathering our team, LPs, portfolio companies, and network of partners across the crypto industry!

[Hasu⚡️🤖 @hasufl]

Thought about this a bit and my take is basically:

Most defi primitives are just finance primitives and have been invented already, just not necessarily built/deployed in Defi yet.

That is because higher-layer primitives require the presence (+ maturity/liquidity/etc) of lower level primitives to become viable.

Thus, technology matters very little for Defi innovation and network effect + ecosystem maturity matters a lot.

To give just one example, interest rate swaps (Pendle) became possible once Ethereum had a flourishing yield ecosystem. That yield ecosystem only became possible when it had deep savers and borrowers, which in turn required mature repo and swap markets, etc. -- Skip any of these steps, and the whole pyramid stops working.

As a result, I expect 90%+ of Defi innovation to continue to occur first on Ethereum and Solana and not any of the new "more powerful" chains. They need to focus on catching up first, paired with focusing on novel areas of apps that need fewer complements to exist.

@HyperliquidX , then, is really a special case among new chains because it has the deepest perps liquidity in Defi, giving it a unique advantage to dominate products where perps are the essential complement/enabler.

The first ones include arguably: volatility and options markets, commodities and certain types of negative carry RWA, and synthetic equity products.

In summary, I expect
- Hyperliquid to take the lead in these areas
- Ethereum (and with some distance, Solana) to continue to lead Defi innovation at large, and
- A more difficult time for new chains because Defi is 90% about network effects and maturity, and little about better technology, and building that takes time.

[Atlas @DentesLeo]

10 years ago I made the decision to never correct anyone who’s wrong unless I’m paid to do that.

Never been happier.

[Hugh Karp 🐢 @HughKarp]

Funds ask @NexusMutual to list protocols for coverage all the time.

A side benefit is the due diligence our team does. Many times funds decide not to allocate instead.

The yield isn't always worth it.

[The Great Martis @great_martis]

Developed countries should brace for 10-15% mortgage rates.

What's concerning is that no technical analyst is discussing the significance of the 32-year break in bond yields.

Even worse: We're entering a multi-decade high-yield era, targeting 7% short-term as deficits and tariffs ignite stagflation. Mortgages? Charging toward 10%+, crushing affordability and debt servicing.

And in a downturn? Yields will dip temporarily before surging as the Fed battles a stagflation-tumbling stock market with QE, pouring fuel on the inflation fire.

Hope this helps.