Arthur Murauskas

This post lives deep, deep inside my brain

@GergelyOrosz IMHO that’s a symptom of a failed organizational structure if none of their very highly paid engineers could prioritize that…

Full postmortem: coinbase.com/en-gb/blog/a-p… I would get this for a small company. But a $40B company? It's a serious WTH moment for me. And my impression of Coinbase engineering nosedived. This is resilliency basics: if you depend on a region, have auto failover and exercise it...

The postmortem from Coinbase's 10-hour outage is out and... damn They run global trading from a single region because of latency. OK, I understand. BUT they have no automated failover prepared! Are they praying the region never goes down?? Doesn't compute for me...

Everything Durov says about the bullshit WhatsApp lawsuit, about France, about "not one byte"—it is all just flim-flam to distract you from the important thing: *end-to-end-encryption*. If you use Signal, WhatsApp, or iMessage, your messages are secure. Not so on Telegram. 🔚

@mitchellh Like if it came straight from the Silicon Valley series “We are making the world a better place by building the emotional layer of trust between humans and AI”

This company literally just sends iMessages as a service.

brb rewriting bun in lisette lisette.run

I strongly believe there are entire companies right now under heavy AI psychosis and its impossible to have rational conversations about it with them. I can't name any specific people because they include personal friends I deeply respect, but I worry about how this plays out. I lived through the great MTBF vs MTTR (mean-time-between-failure vs. mean-time-to-recovery) reckoning of infrastructure during the transition to cloud and cloud automation. All those arguments are rearing their ugly heads again but now its... the whole software development industry (maybe the whole world, really). It's frightening, because the psychosis folks operate under an almost absolute "MTTR is all you need" mentality: "its fine to ship bugs because the agents will fix them so quickly and at a scale humans can't do!" We learned in infrastructure that MTTR is great but you can't yeet resilient systems entirely. The main issue is I don't even know how to bring this up to people I know personally, because bringing this topic up leads to immediately dismissals like "no no, it has full test coverage" or "bug reports are going down" or something, which just don't paint the whole picture. We already learned this lesson once in infrastructure: you can automate yourself into a very resilient catastrophe machine. Systems can appear healthy by local metrics while globally becoming incomprehensible. Bug reports can go down while latent risk explodes. Test coverage can rise while semantic understanding falls. Changes happens so fast that nobody notices the underlying architecture decaying. I worry.

@josevalim Agree 100% but in all fairness, many big enterprises already outstaffed and cut costs of their IT departments so much that what they are left with is not much better than the vibe-coded AI slop...

Some of the responses say "coding agents are great for my web app, therefore software engineering is indeed going away", as if the whole purpose of programming and software engineering is to build web products. Would you trust your medical records to a vibe coded database? Would you board a plane in which its systems are the outcome of software engineering practices fully managed by AI? Would you trust the safety of your servers and your finances to operating systems where handwritten code is not allowed? Would you ship your code using a compiler implemented in a month by a fleet of coding agents?

@josevalim @sasajuric @ElixirConfEU @chris_mccord But did you say “make no mistakes” before launching an agent? 😀

At @ElixirConfEU, @chris_mccord and I set up two races between myself and a coding agent. One was to add a new feature to DBConnection (max_lifetime) and another was to investigate a potential type system regression (links below). For DBConnection, I gave the documentation of how the option would behave. Not only was I faster, the coding agent solution had 4x more LOC than mine (it changed files and data structures that were not necessary for the feature). For the type system, the coding agent could not fix the problem at all. It did recognise it was a type system regression, and it did provide an initial fix in roughly the same time as I did, but the coding agent fix was wrong. After I fed it additional tests (from my own solution), it failed to come up with a fix. I use coding agents daily, I often ship their code as is, but they can still slow me down, they are nowhere close to fixing all problems I tackle in a given week, and are often not up to the standards I expect from my own software.

3-4-2026 ($) •Anthropic’s Skyrocketing Revenue •A Contract Compromise? •Nvidia Earnings stratechery.com/2026/anthropic…

@DrizzleORM Congrats on the amazing news!

As of today, Drizzle has become the most sustainablest ORM on Earth 🚀 PlanetScale hired entire Drizzle core team to work on Drizzle ORM full-time, which is just… WOW! I mean, I can’t believe this is happening and how amazing this is going to be... Let’s address the most important things: ▪︎ The governance of Drizzle does not and will not change ▪︎ Drizzle remains independent and open-source ▪︎ We will now work more full-timer on Drizzle ORM, Kit and Studio ▪︎ Our social media manager(me) will finally have a salary ▪︎ Drizzle v1 is going to be amazing Thank you all for using Drizzle, without you there won’t be 1.2k GitHub issues and 355 PRs, we’d live a happy life, touching grass ❤️

Hey did you accidentally bomb Iraq instead of Iran??? ChatGPT: You’re absolutely right good catch!

The most important skill of an entrepreneur is the ability to make decisions based on incomplete, inconclusive, and frequently contradictory information.

The people who just blindly toss AI shit over a wall onto other humans without using their brain for even a nanosecond deserve shaming. We need to start a public wall of shame for the public identities (not doxing) of these people so we can have bots that just block them. I don't care at all if you do this in your own projects, but when you cross a boundary where another human has to interact with you, its common courtesy to at the very least spend any amount of time at all thinking (the horror).

@ThePrimeagen Understandable for logs and artifacts but why charge per minute then? Feels like a half-thought decision similar to removing cmd + k panel (which they brought back very quickly).

so this must be in response to the cost of storing logs / artifacts? is this the idea? i am trying to gather what is the cost on the github side its all control plane 1. schedule the jobs 2. manage the life cycle 3. display logs 4. store artifacts are there other things?

ok wtf charging for self hosted?

There’s this recurring trope that Europe is overregulated and the US is this sort of free-wheeling world where anything goes. As with everything, the reality is far more nuanced. I used to believe this trope myself… until I actually lived in Europe and experienced it. In Europe, regulation often operates at the collective level.. think healthcare, labor protections, food standards, infrastructure. These regulatory frameworks are heavy by design in that they create stability by increasing broad citizen-level confidence in them actually functioning. But at the individual level, daily life can be far looser. There are playgrounds in Europe that would be illegal in the US due to their “danger.” People rarely wear helmets.. not even toddlers.. on bicycles in many places. Kids climb trees higher and parents barely care or even notice. Farms are open.. kids can climb all over haystack mountains and nobody asks if their farmer is insured. There is a playground in the NL of *literal* piles of discarded shipping pallets and construction debris with rusty jagged nails sticking out everywhere… and little kids climb all over them with hammers connecting random pieces together. One false step and you’re slicing an artery or losing an eye. Yet there is barely any adult supervision, parents don’t care, and nobody is signing any paperwork or waiving liability. We bring American friends there and they literally cannot believe what they’re seeing. And they don’t let their kids. Activities proceed on the assumption that risk is visible, understood, and partly if not mostly your responsibility. Menanwhile… in the US we paradoxically flip this culture. Collectively, we resist broad social regulation writ large. Individually, though, life is wrapped in micro-regulation everywhere… liability waivers, warning labels, signage, insurance restrictions, endless legal disclaimers. Every activity sees to have some paperwork. Everyone is covering for something. This is a cultural thing. The US actually uses the legal system as a cover for social risk-sharing. In much of Europe, the downside of injury or bad luck is partially absorbed by healthcare systems, disability supports, and social insurance. The cost of risk is basically capped for you. The system carries some of the shock. In the US, harm can be financially catastrophic. When something goes wrong, someone has to pay, and courts become the primary mechanism for redistributing that risk after the fact… not “the government.” The you had to layer in contingency-based personal injury law and jury trials, and blaming someone else for your problems becomes economically logical. There’s little downside to suing, meaningful upside if you win, and enormous unpredictability for defendants.. hence why insurance costs have become comically absurd. So what happens…. Businesses respond long before anything reaches court by engineering out risk in daily life… more warnings, more forms, fewer “at your own risk” type playgrounds or other environments. So Europe can feel more regulated on paper… but in actual lived experience that matters to your day to day existenxe, in the US we are often navigating a far narrower acceptable window of risk. In many ways, the US is the most highly regulated place in the entire world, by far, it’s just not “the government” doing the regulating.

@levelsio Similar powerbanks are everywhere in France.

🇨🇳 I found the rentable powerbanks in China One scan to pay with Alipay and then you take the battery and you return it later This would never work in US or Europe because people would just steal the powerbanks

Didn’t hear this heartbreaking exchange towards the end of the video. Am I going to die?”, - asks the child. “No, you’ll be fine”, - says the first responder. “Thank you! Thank you!”, - says the child. Ukrainian children grow up fast. Sometimes within seconds.

Content and Community The old model for content sprung from geographic communities; the new model for content is to be the organizing principle for virtual communities. stratechery.com/2025/content-a…

@nuqs47ng @codestore_tech Pleasure is ours. Thanks for a great library 🚀

Thanks @codestore_tech & oxom.de for your sponsorship! 🩷 May has been a little slow on updates, but we hit a long-awaited milestone: 📈 NPM downloads have passed 500k/week 🙌