Working at Microsoft, I've just now hooked up to Claude Code (my department was not permitted to use it previously), through something called "Agent Maestro", a vscode extension which I guess pipes claude code API requets to our internally hosted Claude models, including Opus 4.6.

I do wonder if there is going to be much of a difference between using Claude Code vs. Copilot CLI when using the same models.

> I do wonder if there is going to be much of a difference between using Claude Code vs. Copilot CLI when using the same models.

I’m also at MS, not (yet?) using Claude Code at work and pondering precisely the same question.

Is this an indictment of OpenAI's models -- that Microsoft has access to through their investment?

We've had both GPT and Claude models available to us in Github Copilot for some time. At first, it was only GPT models.

> I honestly don’t think the models are as important as people tend to believe.

I tend to disagree. While I don't see meaningful _reasoning power_ between frontier models, I do see differences in the way they interact with my prompts.

I use exclusively Anthropic models because my interactions with GPT are annoying:

- Sonnet/Opus behave like a mix of a diligent intern, or a peer. It does the work, doesn't talk too much, gives answers, etc.

- GPT is overly chatty, it borderline calls me "bro", tend to brush issues I raise "it should be good enough for general use", etc.

- I find that GPT hardly ever steps back when diagnosing issues. It picks a possible cause, and enters a rabbit hole of increasingly hacky / spurious solutions. Opus/Sonnet is often to step back when the complexity increases too much, and dig an alternative.

- I find Opus/Sonnet to be "lazy" recently. Instead of systematically doing an accurate search before answering, it tries to "guess", and I have to spot it and directly tell it to "search for the precise specification and do not guess". Often it would tell me "you should do this and that", and I have to tell it "no, you do it". I wonder if it was done to reduce the number of web searches or compute that it uses unless the user explicitly asks.

Compare their system prompts and the agent harness logic. It's 99% of what makes the agent useful, and it can be quite different.

Crowding around our first ever computer, a 120mhz pentium with 16mb of RAM and a 1.6gb hard disk, watching that Weezer video on the CRT monitor with my whole family is a cherished memory.

ya, starting with M1, Apple really broke away hard. Windows laptops, especially of the thin-and-light variety, are so bad by comparison.

Very cool, I have not given Ardour a try in many years but the UI looks quite nice. Will give it a go

Maybe at a greenfield startup. Where I work this idea wouldn't be entertained for a millisecond.

This does make a quite a bit of sense. When I was a teenager in the 90s/early aughts, it was all IRC, script kiddie stuff. Reckless abandon. What worries me is that it seems like full-grown adults are happy to accelerate the dead internet and put security at risk. I assume it's not just teenagers running these stupid LLM bots.

I don't think "we" would have been impacted since this specifically targets the updates, but recently Microsoft pulled Notepad++ from the list of apps we can use on our production management laptops. Some people were annoyed and whining about this. That predated this announcement by a few weeks. Probably the right move by the security folks.

it was pulled because the binaries were self-signed for a short period, not because they knew something

who signed the binaries was irrelevant for this attack, because the issue was not checking any signature

nice score! I got 129, and yes, same experience.

129 here, not bad for the end of the day. I listed a surprising number of dinosaurs, and of course, edible animals, and Lion King stuff.

Warthog! Meerkat!

AWS accounts for more than 50% of Amazon net profit though