I disagree, somewhat.

Many SAY they want elite engineers.

But what they really want is someone who can ship the worst possible muck as fast as inhumanly possible, and still manage to help keep the lawsuits down to a dull roar of just multiple billions of dollars per year. And they want to pay these developers literally zero.

Which is why they all thought that tokenmaxxing would be such a good thing, since it could be used to build virtual engineers that literally cost them absolutely no money at all.

Yes, I worked at Amazon. I know that culture. And they had some seriously elite engineers, at least when I started there.

I also did a six-month stint at Apple Retail Software Engineering a few years earlier. They also had some seriously elite engineers there. Honest-to-goodness real-deal 10x types. And they frequently got yanked from one project to another based on whatever the whim was of Steve Jobs, or whomever was in charge. And they were constantly being pushed to actually produce 100x, not just 10x.

At the scale large enterprises operate this is quite rational. It's simply not feasible to recruit and retain enough elite engineers to meet their goals. So they have to design org structures and processes to extract useful work from mediocre engineers.

An “elite engineer” is only worth their cost when they are given agency and allowed to chose what to work on.

Most executives(as in the executives at the majority of companies with valuations below big tech) would rather kill themselves than let an employee have any agency.

Seeing lots of content on the web about pangram's unreliability

Is there anything this dude writes that isn't AI generated. I guess you'll have to go back to his 2021 and before blogs.

Isn’t this the tool that flagged original text from Mary Shelley’s Frankenstein as “100% AI”?

> maybe not in their first 10 years of their career, but what companies do is cut that short and push engineers away from the technical work

Perhaps the issue is that there just isn't that much technical work to go around.

While software was booming I remember hearing stories from EEs who found the opportunities in that field lacking. My understanding is that a lot of the people didn't do much technical or engineering work, but those who remained in the field became "BOM spreadsheet babysitters" and "Guy who's only here because we need a licensed engineer to sign off". It really pushed my away from that field even those there are many domains within it that greatly interest me. My understanding is that a lot of the hands on or technical engineering positions went to Asia along with the manufacturing

...and by the way AI isn't any of the 5000!

I think you're right that many more people are in theory capable of joining the group, but not that the industry is conspiring to keep it small. If anything it's the people who have an academic history that stretches back to childhood that end up dependent on having someone to guide and grade them. Beating your own path through the wild is scary, uncertain, and often thankless, so I think you'd only want to do it if you already felt sure you were among the most motivated and dedicated members of a group.

I mean you could just as easily claim there are only 5000 “elite” software roles available at any given time.

In other news, people believe the market is fair and the best people get paid the most.

We all know that people in Africa work much much harder than your average western person, and yet they get dollars a day while you and me get 100x more.

Life isnt fair and to think that the best always win is plainly wrong, because we both know those poor people in Africa deserve far more than they get.

this is for part of it. I have been SWE for almost 30 years now and do not have a resume. I interviewed for a ton of positions when I graduated from Uni, got my first job and subsequently never looked for a job again.

I think more of an issue here is that there is just way too many SWEs and we can’t all be “elite.” In my 30 years in I have worked with roughly same percentage of “elite” as the post is alluding to. The history/structure of just about every project I worked on boils down to needing two distinct group of SWEs, the inovators and closers. The inovators are the “elite” (within the project) and they figure all the shit out. The closers crank it out once all the shit is figured out for them

Most of the true elite have spent years working anonymously with no publicity. All of their work is locked up behind an enterprise firewall. They never post on social media or contribute to open source or speak at conferences.

That's the list.