BiraIgnacio

Thanks so much for the work you and your team do!

One could say in order to be a senior developer in any area, more-than-good- communication skills are required.

I dislike Go but I have to admit, it's a great language for AI generated code. Simple enough, it compiles quickly and it performs meh-well enough for most applications.

One of the reasons I dislike Go is because it's easy for most engineers to write really low grade code with it. But AI agents would probably not write the best code in any language anyway, so not much is lost.

I'm sure someone at Anthropic is reading this post. I hope so and I also hope they take the high road.

I like to think the Vision Pro was ahead of its time. Just because it's such a cool product. Just not practical yet for every day use cases

The hacking aspect has been hit and miss for me. Just today I was trying to verify a fix for a CVE and even giving the agent the CVE description + details on how to exploit it and the code that fixed it, it couldn't write the exploit code correctly.

Not to say it's not super useful, as we can see in the article

True, it's better than most for sure and I agree it used to be better. Though a lot of other software for windows and linux are really not that great so the bar is probably on the lower end.

Does anyone know if these or similar techniques are still used in today's suits production?

love it!

Nice work and thanks for sharing it!

Now, I ask, have LLMs ben demystified to you? :D

I am still impressed how much (for the most part) trivial statistics and a lot of compute can do.

I never got OpenClaw and I'm unsure what this means exactly. Would it be just that you can't put your Anthropic API Key/credentials in OpenClaw and use it? Sounds like something easy to bypass, if it's just that.

If C++29 was exclusively about quality-of-life improvements, improving what exists, I'd bet the community wouldn't mind too much.

my absolutely-non-expert guess is that it would work much like any other fuel? Combine with matter, get a lot of head out of it and use that in the best way we know.

There's an interesting bit about talking to LLMs to help dispel conspiracy theory ideas. I can see how that would work but also, how that could backfire as well.

Great podcast episode in any case.

I'm happy to see so many philosophy or philosophy-adjacent books on that list. And I also wonder why that is.

I ran into a similar case recently, there was a ticket describing what needed to be done in detail. I was written by a human and it was a well written spec. The problem is that it removed some access controls in the system, essentially given some users more access than they should have.

The ticket was given to an LLM, the code written. Luckily the engineer working on it noticed the discrepancy at some point and was able to call it out.

Scrutinizing specs is always needed, no matter what.

There's nothing wrong in being nice and some chit-chat. Any kind of work, well most kinds of work, are about people and relationships. Building something with people when people can't relate to one another is quite hard.

Open source book dedicated to helping you to make the best possible sourdough bread at home.

Oh thanks, this is a pretty good elevator pitch, now I see the connections I was missing.

That's a very sensible, realistic and non-BS response.

I'm glad to see this coming from a company that is so popular these days.

Thanks!

One of the down sides of Vibe-Coded-Everything, that I am seeing, is reinforcing the "just make it look good" culture. Just create the feature that the user wants and move on. It doesn't matter if next time you need to fix a typo on that feature it will cost 10x as much as it should.

That has always been a problem in software shops. Now it might be even more frequent because of LLMs' ubiquity.

Maybe that's how it should be, maybe not. I don't really know. I was once told by people in the video game industry that games were usually buggy because they were short lived. Not sure if I truly buy that but if anything vibe coded becomes throw away, I wouldn't be surprised.

> It was the fast food of the chōnin class in the Edo period.

I definitely see that with C++ code Not so easy to "fix", though. Or so I think. But I do hope still, as more and more "modern" C++ code gets published

It is joyful. I think of it like a proof being written and that is confirmed to be correct (most of the times) in the end.

> shipping a product that solves real problems for real people.

Off-topic but I must... Love to see that there are still software companies doing that! Really great work. I'm not an expert but the product looks great.

hypergrowth is very hard. First to be able to get there and then, once there, to keep up offering quality services.

Finance is not my thing so this is baffling to me. Unity stock is/was down 20% at some point? if it's just because of this, it's some reaction.

Great content and approach, thank you!

Ursula's notes really enrich the work. Fantastic ways to render insights in words.

oh I've been looking into whisper and vosk in the last few days. I'll probably go with whisper (with whisper.cpp) but has anyone compared it to vosk models?