Wazzymandias

I don't need a more bloated operating system or greater resource usage for things I don't want, use, or need. i'd prefer a calendar app that doesn't suck, and an OS that doesn't decrease performance by 5-15% on every new upgrade for slightly older machines

It's amusing that this blog post explained Jev's underlying design far more clearly than their onslaught of bull posts and marketing

This entire time they could have just said "decision model" but they kept using vague, flowery wording. I have no idea why

this is awkward - the site definitely still looks vibe coded / LLMs helped make it

> Why check it? It's obviously going to be correct.

Because as complexity floor increases, it's "going to be correct" in hyper-specific, hyper-literal, insidious ways, with 10-50x more lines of code than necessary, and tens to hundreds of incredibly useless tests that give the illusion of quality, and cause cascading effects where seemingly irrelevant and orthogonal features that were once working end up breaking because of the agent's changes

> been trained on enough long context tasks to have become pretty good at planning

This is absolutely not true

depends on the blast radius of zero-days, it's not like there's a continuous immediate release process for these models; they can eval internally before releasing publicly

This looks and feels a lot like productionized conformal prediction

Similar reasons why Israel hasn't experienced any sanctions yet despite their ongoing genocide against the Palestinians

It's not a conflict, it's a genocide

Some good takes, some bad takes in the post. The analogy of cutting a tree is pretty dehumanizing and doesn't factor in the morale and culture hit laying people off does to a company.

Sustainable growth is far better than boom and bust cycles. I guess if you're shutting down entire product verticals you can reduce the blast radius. Doing partial cuts across the board can cause top talent to look elsewhere.

It depends on the definition of "relatively simple features"

Mismanaged startups can have a lot of technical debt as a result of pivots, shifting priorities, weak engineering and an emphasis on being feature factories. As a result you can get to a point where even a relatively simple feature takes months because of massive technical debt.

Shipping early and often while giving engineers the space to tackle debt is important. Good hiring pipeline for competent product focused engineers can help mitigate building castles in the sky. That and/or having a good CTO.

It's been mostly from personal experience, though I shouldn't generalize to the community as a whole.

Anecdotally my experience with nix community has been unpleasant. They're very resistant to suggestions and offer little in terms of guidance or documentation.

Nothing should be labeled as the ultimate devops toolkit when its documentation is as atrocious as nix

People can commend it as much as they want, but the steep learning curve is largely self inflicted because of their resistance to writing clean, comprehensive, up to date docs

It has also led to the community being filled with a lot of arrogance and pretentiousness

I wouldn't run nix in production because of the lack of accessibility and toxic community. There are other ways to get reproducibility, etc without using an arcane and poorly documented toolkit

Would anyone happen to have an ELI5 of how the M1 chip achieves greater power efficiency?

I think the problem is the industry standards and expectations for candidates. The fact that OP links an archive of interview questions is indicative of the underlying problem.

Giving a candidate a problem they haven't seen before and observing their problem solving skills and mentality doesn't sound so bad on paper. However the process has been gamified and as a result the expectations for what a candidate should know and accomplish have increased. This ends up favoring younger candidates and/or fresh grads, since on average they can expend greater time and energy into studying for these types of interviews and their algorithms & data structure courses are still fresh in their minds. A more senior candidate or a person with more responsibilities outside of work (taking care of children, parents, etc) won't be able to invest as much time into studying these types of problems. The original purpose of the interview is also obfuscated - did the candidate genuinely show good problem solving skills or did they see the question before and merely memorized the solution? As a hiring manager I've seen multiple instances of the latter.

I think the utility of spending time on these problems varies from person to person. I personally find it more useful to delve into distributed systems, cryptopgraphy, networking, or operating systems. These topics have made me a much better engineer and improved my problem solving skills, more so than programming riddles which gave me diminishing returns over time.

Apologies if these are naive questions, but why does Microsoft require use of Edge to track user data? Can't they obtain data from the DNS requests any browser makes to access websites?

And are there any ways to mitigate the amount of data they obtain when using Edge (e.g. Ublock)?

I've tried so many options for note taking but I can never find a good solution that fits all my needs. I find writing things down helps with comprehension and retention, but using a physical notebook has its limitations when you want to reference images or snippets found online.

In the end I landed on OneNote, but I find it to be slow for larger notebooks, and it doesn't have an intuititive method of linking different notes from different sections for easy reference. However I like the handwriting recognition far better than Apple variants, and I felt like with apps like Notion I spent more time customizing and grokking the app than actually taking notes.

"It could be he has been busy doing real work in the last ten years and not researching solved problems that are generally encapsulated in well tested existing modules."

Sorry, I should have been more clear. We didn't ask the problem with the expectation of a perfect solution, it was more to gauge his problem solving skills. However as we were running out of time we asked him to implement a naive solution, which was met with a very negative response. Maybe I'm not explaining it properly but it gave the impression of a serious ego problem. Actually part of the criteria for our interview is to see how well a candidate can take feedback and make changes accordingly. They did poorly on that front.

Agree that architecture leadership is critical, especially when dealing with business teams that acquiesce to every single client demand with impunity.

In my previous job we hired someone in a "Senior Architect" position. They had extensive experience at a larger company, but there were red flags from the beginning (including the interview, where they got visibly upset and melodramatic when they were unable to finish the algorithmic problem we gave)

Working on anything related to design and architecture with them was a disaster. They preferred waterfall over agile, didn't take feedback very well, was obsequious towards leadership but toxic towards the team they led, and would terminate collaboration with someone if there was disagreement.

The soft skills for an architect are essential, and I think that includes excellent communication skills, documentation, and ability to take feedback and constructive criticism. Needless to say, the coworker I describe fit none of those criteria. It's been pure bliss not having to work with someone like that, and helped me better understand what to avoid in the path to becoming a legitimate software architect.

"Islam was out because that’s not for white people."

For someone who espouses an expertise in mathematics his perspective on religion is quite reductive if not completely wrong in some cases. I guess in addition to math the author is fond of strawman arguments.

Deletion can also be a little misleading.

I deleted (not deactivated) a Facebook account that I had from high school to sophomore year of college.

A year or two later I made a new one and didn't fill out anything except my name and didn't add any friends (yet). Almost immediately facebook started suggesting friends, groups, etc that existed on my deleted account.

tl;dr nothing is every really "deleted," just hidden

That's disappointing to hear but thank you for the honest response. I currently work in Ruby but my favorite language is C, I feel like it helped me become a better developer and improved my understanding of imperative languages. Sucks that something so useful is cast aside for something like PHP.

Thank you! Your explanation made the most sense, a lot of the other ones went over my head. This is so clever and fascinating.

I honestly have no idea what's going on, but his writing is humorous.

Whiteboard coding isn't really a metric close to what is required in the job though.

I like to write my algorithms on paper first before writing code, but doing it on a whiteboard in 45 minutes while explaining my work and checking for edge cases can be a bit unreasonable.

I don't think the contention is about testing a programmer's ability, but the way their ability is tested. Personally I think pair programming is a better method of testing someone's ability to code AND their ability to collaborate with others.

What I think about when I use Facebook: don't use Facebook.

I very much agree with this, but just as an aside I'd like to mention that I worked on a full stack web application as a side project and I believe that helped me land my current job at a startup.

However I'd also note that I applied to a LOT of companies, and while my side project helped me land my current job, 90%+ of the companies didn't care about it.

I learned a lot from my project but job hunting is a numbers game, and it's far more worthwhile to study than it is to build a side project IMO.

Thank you for the comprehensive response! I have a lot to learn but it's exciting.

What would you recommend instead for a beginner trying to get into ML?