Purtle (@knwlim) header
Purtle (@knwlim) profile picture

Decentralized, transparent, permissionless, programmable, trustless, censorship-resistant, unbanked, self-sovereign degen

Joined July 2009
Photo by Purtle (@knwlim)Photo by Purtle (@knwlim)Photo by Purtle (@knwlim)Photo by Purtle (@knwlim)Photo by Purtle (@knwlim)Photo by Purtle (@knwlim)Photo by Purtle (@knwlim)Photo by Purtle (@knwlim)Photo by Purtle (@knwlim)Photo by Purtle (@knwlim)
Purtle retweeted
imagine dodging north korean spearphishing, home invasions, defi vault contagion, ledger supply chain attacks, ai driven cryptanalysis breakthroughs, quantum computing threats and more just to lose 20% from price of majors going down.
6
1
2
88
5,865
A new piece in the Financial Times created a great visualization of the results from this study. TL;DR: Rent control helps people whose rents get limited, but it hurts everyone else. It pushes up rents, pushes down new building, and makes cities worse. It is horrible policy.
What do studies say about "freezing the rent"? Let's have a thread🧵 First thing's first: Most studies agree that rent controlled units have lower rents, but also the supply of rentable units goes down and un-controlled units see their rents increase. Uh-oh!
98
761
69
5,371
192,339
Purtle retweeted
Where does money invested into the AI buildout actually go? For every $100 flowing into the supply chain: - $50 to chips - $20 to power - $15 to networking - $15 to cooling, buildings, and land More charts in State of Markets II: a16z.news/p/state-of-markets…
Introducing our State of Markets pt 2, along with a companion podcast where we unpack the data and discuss what comes next. Tech is the everything cycle. Supply: putting the buildout in context, just passed railroads as % of GDP. The wisdom of Elon is real: the factory (or the datacenter!) is the product. Demand: diffusion is so, so early. Median AI vendor spending in the top 1% of companies is 8x that of the top 10%. Only about 30% of S&P 500 companies report a quantified AI impact, which means there’s a substantial opportunity in connecting models to a company's data and workflows. Diffusion into companies is one of the main themes of the next 5 years. We’re entering the agent work period. Only a few million users today, but applicable to billions of internet users with massive surplus created. META/GOOG monetize US users at $200+ per year today. Agent opportunity is much higher. Mega-trends the next 5 years: Consumer agents, Robotics, Autonomy, AI x bio, Personal health, Diffusion into enterprise, New era of American Dynamism. Much more in our SoM report here - a16z.news/p/state-of-markets… @a16z @sarahdingwang @aleximm @santiago__rdz
177
869
154
4,686
1,063,555
Purtle retweeted
AI will be especially fatal to writing that takes too long to get to the point. Now there will always be a more concise alternative, and 99% of readers will choose it.
275
95
52
2,343
198,345
Purtle retweeted
sex is great but have you ever spent the whole bear market buying shit nobody wanted then suddenly 3 green days erase months of suffering and put you deep in profit?
92
204
17
4,352
141,476
HYPERLIQUID
[🌲] TRUMP: SEC CHAIR WORKING TO BRING HYPERLIQUID INTO US
1
5
50
Purtle retweeted
The chart of housing affordability is almost a direct inverse of how many new homes are built It's that simple Build more houses!
100
76
32
1,137
416,423
New from me: Until recently, the gap between comfortably-off and just-getting-started could plausibly be crossed in a decade or two of hard work. Growth in asset prices means that’s no longer true. My column on moving from income world to wealth world: ft.com/content/6a35508c-c0bd…
87
429
113
2,409
932,954
Purtle retweeted
I’ve wanted to tell this story for years. Never had the courage. Here it is. I turned a presale allocation into $80 million on $OHM. Today I have $500k left. In 2021, I got a presale allocation in OlympusDAO, then aped heavily myself on top of it. The allocation got me in the door. My own conviction made me go all in. Staked everything. Watched it compound daily. By the peak I was sitting on $80 million. Then I started spending like the money printed itself. Private jets to Dubai because commercial felt beneath me. $40k weekends in Monaco. A garage full of cars I drove twice. Watches I never wore. I tipped $5k at dinners just to feel something. Every purchase was a flex for an audience that didn’t care. The casino was worse. High limit rooms in Vegas and Macau. I’d lose $2 million in a night and laugh it off because the portfolio would make it back by morning. Until it didn’t. When $OHM unwound, I didn’t sell. I doubled down. Then I leveraged. 5x, then 10x, trying to trade my way back to the peak. Every liquidation felt like a personal insult, so I’d open a bigger position. I wasn’t trading anymore. I was gambling with a different interface. $80 million became $20 million. $20 million became $4 million. I told myself $4 million was still life changing money. Then I levered that too. $500k. That’s what’s left. Here’s what I learned the expensive way: Unrealized gains are not money. I never had $80 million. I had a number on a screen and the arrogance to believe it was permanent. Getting in early is a gift. I treated it like a skill. The allocation didn’t make me a genius. It made me lucky. I confused the two for three years. Lifestyle inflation is a leak you don’t notice until the ship is underwater. The jets and cars didn’t kill me. The identity did. I became someone who needed to spend to feel like a winner. Leverage doesn’t get you back to even. It gets you to zero faster. Revenge trading is just grief with a chart open. Nobody at the table in Monaco remembers my name. I’ve carried this story alone for years. Too embarrassed to say it out loud. But $500k is more than most people will ever hold at once, and I’m done pretending the past didn’t happen. The next decade is about building slow and keeping what I make. If you’re up big right now, screenshot this. You’ll need it.
366
213
83
3,522
577,450
Purtle retweeted
1. Crypto crashing despite rising dollar anxiety 2. Can no longer say 'we're so early' 3. Crypto Twitter is dead 4. Institutional adoption already happened, tailwind spent 5. Regulatory environment is already as good as it gets 6. AI crowding out electricity for miners 7. Crypto is Epstein-adjacent (Epstein Files) 8. Quantum computing threat to BTC's security model 9. DATs like Strategy are now sellers 10. AI taking all the smart-tech mental share 11. Other markets pumping hard (SK Hynix +250%, Micron +260% YTD) 12. Everything rally is back, Goldman non-profitable tech basket mooning like 2021 Yep, I feel it all.
12 REASONS WHY IT'S THE COLDEST CRYPTO WINTER EVER Back in February I wrote a list of 10 reasons why this the worst crypto winter ever. Well everything I cited then still holds, but now I have 2 more ways it's gotten worse: From the newsletter bloomberg.com/account/newsle…
37
34
4
374
82,887
This is a critical factor to consider. Great piece by @jburnmurdoch @ft What if remote working, not AI, is to blame for weak junior hiring? ft.com/content/2205e2d0-50dc…
11
42
7
190
36,873
Purtle retweeted
College CS enrollment is declining Charts of the Week: a16z.news/p/charts-of-the-we…
43
62
37
623
399,393
Purtle retweeted
Ken Griffin went home on a Friday "fairly depressed" after watching AI agents at Citadel do work that used to take teams of PhDs in finance months to complete. Done in days. His words: "These are not mid-tier white collar jobs. These are extraordinarily high skilled jobs being automated by agentic AI." This is the head of one of the most successful hedge funds in history saying the people he pays seven figures to analyze markets and structure deals are being replaced by software that works in hours instead of months. Not theoretically. In his own office. Right now. The Coatue deck we covered earlier this week called agents "the biggest unlock" in AI. Griffin just confirmed it from the buy side. The shift from copilots to agents is not a future event. It is already happening at the highest levels of finance.
698
1,733
456
12,203
6,525,944
Purtle retweeted
Crypto's current state is a bit shit 1. Market cap is not an indicator of quality - the top 50 is made up of ghost coins or bloated governance slop that has underperformed and is uninvestable 2. The long tail speculative stuff went from high risk high reward to 'some dude in Miami is going to zero this if you hold it for more than 5.9 seconds' 3. Everything is extremely correlated and you can't meaningfully make bets based on sectors as it all converges into a tightly correlated mush, especially to the downside 4. Broad brush alt season is an artefact of the past that's very hard to replicate given (2) and given that there are simply too many coins and the excess of speculation doesn't really happen on centralised exchanges anymore - it's been siphoned off to bundled shit in max PvP settings 5. Crypto reputationally is no longer the sexy frontier of speculation. Institutional bid is in AI, retail speculative bid is in 0DTE equities, single name stocks etc. 6. Convexity has flattened. Even a lot of the historically safe blue chip stuff (BTC, ETH etc.) has underperformed and the historical anchor of 'buy deep drawdowns because all-time highs are guaranteed and explosive' has disappointed. All the shit we used to put up with because of the accessibly massive trend and momentum effects is now harder to justify because those same effects are getting neutered or siphoned off into other arenas. The obvious rebuttal is 'cycles' but even this past cycle is a useful counterpoint: it was extremely concentrated versus broad brush wealth effect, plus something very obviously broke after 10/10. So what does this all mean? 1. In previous cycles, nailing timing was enough and selection was the cherry on top (rising tide lifted all boats). I don't think that holds - both timing and selection matter now and in the future. 2. Participation alone can be an edge if the asset class is early enough and/or mispriced enough. I don't think that holds either, and we might actually have to learn how to trade (fuck). 3. Hopefully I'm an idiot doomposting the bottom GM
205
182
40
1,761
173,440
I’ve never seen a map that makes more sense.
380
2,133
287
17,568
1,703,193
AI increases workload. Many such cases.
"Using coding agents well is taking every inch of my 25 years of experience as a software engineer, and it is mentally exhausting. I can fire up four agents in parallel and have them work on four different problems, and by 11am I am wiped out for the day. There is a limit on human cognition. Even if you're not reviewing everything they're doing, how much you can hold in your head at one time. There's a sort of personal skill that we have to learn, which is finding our new limits. What is a responsible way for us to not burn out, and for us to use the time that we have?" @simonw
128
68
15
1,091
199,929
Purtle retweeted
i expanded this eval to 25 languages and evaluated gpt-5.4 medium /fast in codex you can see the full results below github.com/banteg/minigit-be…
I benchmarked which language works best with Claude Code. Ruby, Python, and JavaScript came out cheapest, fastest, and most stable! See my article in detail: dev.to/mame/which-programmin…
13
14
2
131
21,299
Purtle retweeted
Introducing EVMbench—a new benchmark that measures how well AI agents can detect, exploit, and patch high-severity smart contract vulnerabilities. openai.com/index/introducing…
1,136
1,252
845
8,697
2,618,025
AI benchmarks are losing their relevance Solving abstract puzzles or passing theoretical exams doesn't prove real understanding At some point we're going to hit a wall with these artificial tests and realize that assessing human intelligence is harder than we thought
6
44