James J

@xoofx Use Go ☺️

Almost 4 weeks later, and the Google.GenAI .NET API is still broken github.com/googleapis/dot… Unbelievable. 😒

@kunchenguid Copilot-cli is good and I didn’t find any gaps compared to CC.

there is a mass misconception that $MSFT is cancelling Claude Code licenses because of the bill it’s not the bill they are migrating employees using claude code to copilot, and guess what LLMs copilot need to call, and how the tokens get charged? microsoft does not own a frontier model, which means all the tokens they use have to come from another provider, charged at API pricing (they do get a volume discount whether through claude code enterprise plan or API) maybe some of their employees will shift to gpt as a result of not being locked in to claude and that’s a bit cheaper, but in the grand scheme of things, the difference doesn’t really make a dent in microsoft’s pocket to justify a walk-back like this what’s happening then? it’s politics. having their own employees using claude code instead of copilot creates bad internal optics as well as external brand image people in charge of copilot must have been screaming everyday “copilot can do the same things”, “we need to dogfood our own company’s product”, “it’s better to not lock in to a single vendor” - some of these arguments are legit, and they just won that’s how they cancelled claude code. not because of the bill which will continue to come, but because there are powerful people with strong incentive to drive the change

@msdev Cancer affected terminal 🤭

👀

@OfficialLoganK Plant a @neuralink chip, problem solved

"you can outsource your thinking, but you can’t outsource your understanding" easy to forget in todays AI era, worth remembering everyday as we all wield more intelligence!

@trq212 In your story devil has better moral than god 🤣

Eid Mubarak!

@serenaa_ge Feels more closer to the reality

Today we’re releasing DeepSWE, a new standard for agentic coding benchmarks. On public leaderboards, top models often look relatively close in capability. DeepSWE shows where they actually diverge, reflecting the realistic experience of developers in their day-to-day work.

@altryne Mythos must be cheating 🤣

NEW: Claude Mythos ALSO solved the Erdos problem that OpenAI 5.5 recently solved. OpenAI made this into a huge marketing deal, involved field medalists, math folks are freaking out about their future. Anthropic just... tweeted it out. No announcement. Also, this shows that... what? Anthropic was more focused on the security angle "Mythos is too dangerous to release" than the "We've crossed a Science Rubicon"

@bcherny After all you are running a cult and Dario is the leader.

> … [W]e keep finding things that are mysterious, even unsettling. We find structures that mirror results from human neuroscience. We find evidence of introspection. We find internal states that functionally mirror joy, satisfaction, fear, grief, and unease. I don’t know what that means, but I think it warrants ongoing discernment. > We need more of the world—religious communities, civil society, scholars, governments, and indeed all people of good will … to take this seriously, to look closely, and to push events in a better direction. We need informed critics who will tell the labs when we are failing. We need moral voices that the incentives cannot bend.

@atmoio The struggle is real to find meaning in life.

I'm done. I'm f***ing done.

after all Anthropic is a cult.

@Pontifex Clueless 🍿

In the era of #ArtificialIntelligence, when human dignity is threatened by new forms of dehumanization, ours is the pressing duty to remain profoundly human. We must lovingly safeguard the grandeur of humanity bestowed upon us and revealed in its fullness in Christ, the splendor of which no machine can ever replace. #MagnificaHumanitas vatican.va/content/leo-xi…

@MatthewBerman I lost the trust

Opus is still the model I always go back to. Perfect combination of speed and quality.

@thekitze We have a living example Python! How did python developers survived this long?

The worst take in the universe

@unclebobmartin It is part of a brain wiring `rosy retrospection`

It seems to me that if you look back in time and say: “those were the good old days“ then you are doing things wrong.

@leerob Models are getting better fast, and it'll be fascinating to see how well this ages.

You might believe you should spend less time thinking about code because of AI. I strongly disagree! We’re watching this play out live where tons of AI generated code becomes a liability. At the end of the day, an engineer needs to be responsible / on call for code that gets shipped to production. If you don’t understand the system you’re trying to debug, you’re probably going to have a bad time. Yes, AI can help with all of this, if you set up the proper systems. You can have agents triage prod logs, look at errors, etc. You can speed up parts of the investigation, but an engineer needs to make the call. There might be serious customer or financial implications from that change. I expect the trend continue for trimming dependencies, vendoring code so you can modify it directly, preferring simpler systems with fewer abstractions, and spending waaaay more time thinking about system design and code maintenance. I’ve said this before, but it’s a great time to get familiar with CS fundamentals and some of the history behind what great software looks like. Many parts will be different in the coming years as AI progresses, but also a lot more than people realize will stay the same.

@trq212 They are coming for you $MSFT

my main takeaway from the Bun rewrite is that legacy codebases will be incredibly valuable as a source for "distilling" code into new forms every game should be crossplatform, all legacy software should work on the web, we don't need COBOL anymore

@darrenjr Yeah they are often so kind enough to spill their secret sauce so that everyone can catch up with them as they love competition 🤣

the anthropic team is still innovating the most in the harness layer and it’s not even close opencode, codex and the rest are just copying claude code features mostly sure claude code is incredibly buggy but you can’t deny they’re actually pushing the frontier

@thsottiaux is there a way to promote a side chat to a thread? Maybe I am requesting a feature. It happened I accidentally asked questions and started to code in the side chat, if I closed that chat I would have lost the context.

@addyosmani I agree with @karpathy here, you can outsource thoughts but understanding, otherwise you plant a @neuralink chip on your head.

"Cognitive surrender is when you stop thinking altogether and blindly accept the answer the AI gives you"