I guess we see things differently.

They don't need to be especially talented engineers, but, in my experience (and I actually have quite a bit of it, in this area), they need to be dedicated to a culture of Quality.

And it is entirely possible for very talented engineers to produce shite. I've seen exactly that.

A culture of quality doesn’t require particularly skilled individuals to function.

That’s in fact the thesis for the entire Deming management philosophy, and in line with what I’m saying (you can produce high quality with a good process or a good culture, you don’t necessarily need high caliber individuals)

Japan used a lot of Deming’s theories, to significant success. I worked for a Japanese company.

In my case, the company produced absolutely top-shelf stuff, but even relatively mediocre companies did well, using Deming’s techniques. It required that everyone be on board, wrt the culture, though.

But I have found that a “good” engineer is one that takes their vocation seriously. They may not be that accomplished or skilled, but they have self-discipline, humility, and structure.

I’ve met quite a few highly-skilled “not-good” engineers, in my day. I’m embarrassed to say that I’ve hired some of them.

> you don't need "high quality engineers" to have high-quality outputs.

[citation needed]

37signals and craigslist are two examples (and they’re open about their engineering being sub-par). If you consider FB products “high quality”, it’s another example (the average FB developer is anything but “high quality”, by most definitions). Palantir is another example, with a horde of junior engineers and famous for bad practices (yet here they are commanding the US military). And so on and so on. The inverse is also true - plenty of stellar teams producing irrelevant or low impact products.

FB products are low quality as is 37 signals. I don't think you necessarily need geniuses to build good products but your examples are bad.

The article and GP mentions this as well in a roundabout fashion; a high-quality engineer is a waste if the organization around it fails. It's better to have mediocre developers in a mature organization than a hero developer working in the shadows. I've seen a few.