5 ms·

Maybe. I have worked at both startups and FAANGs. There are developers with CS degrees who are clearly worse than those without. But I could always tell who d

Maybe. I have worked at both startups and FAANGs. There are developers with CS degrees who are clearly worse than those without. But I could always tell who did not have a CS degree, based on poor understanding of things like running time analysis, good systems design, more complex algorithms etc. The best developers always had at least a CS undergrad. Though there are clearly diminishing returns with going above that, unless you are a researcher. Still, the goal of this site is commendable. There are tons of folks who don't have degrees because they could not afford it.

mnm1 7y ago
Many people with CS degrees don't understand those topics and many people without a CS degree do. So I don't buy this.
- codesushi42 7y ago
  Well, that's what I said. But non CS folk never did understand, in my > 10 year experience. Of course there will always be exceptions. John Carmack is one of the greatest programmers of all time, and he never went to college. To be fair, he did excel a lot further under the mentorship of Michael Abrash, who was a researcher in 3d graphics. But for your average Joe schmoe, this is likely not to be the case. Would I not hire someone on the basis of no CS degree? No. Would I be more skeptical? Absolutely. Would I consider them for a senior advisor role? Probably not. Unless it's from another technical field (math, physics). Even then, I don't know, it depends on the role. I've seen a lot of shit code from people who were mathematically brilliant.
- sokoloff 7y ago
  I've found it much more common for a non-CS degree holder to start down a path that boils down to "first, I'll solve the halting problem and then I'll apply that to the problem I'm actually facing" or similar foundational errors. (I say this as someone who has had a 26 year career in software without holding a CS degree.)
  - souprock 7y ago
    If you aren't demanding 100% reliable success, "first, I'll solve the halting problem and then I'll apply that to the problem I'm actually facing" is a viable strategy. My workplace does this, especially with the PhD holders. It's kind of amusing to put aside well-founded fears about the halting problem. If you are clever, it works more often than most people would imagine.

mnm1 7y ago

Many people with CS degrees don't understand those topics and many people without a CS degree do. So I don't buy this.

codesushi42 7y ago
Well, that's what I said. But non CS folk never did understand, in my > 10 year experience. Of course there will always be exceptions. John Carmack is one of the greatest programmers of all time, and he never went to college. To be fair, he did excel a lot further under the mentorship of Michael Abrash, who was a researcher in 3d graphics. But for your average Joe schmoe, this is likely not to be the case. Would I not hire someone on the basis of no CS degree? No. Would I be more skeptical? Absolutely. Would I consider them for a senior advisor role? Probably not. Unless it's from another technical field (math, physics). Even then, I don't know, it depends on the role. I've seen a lot of shit code from people who were mathematically brilliant.
sokoloff 7y ago
I've found it much more common for a non-CS degree holder to start down a path that boils down to "first, I'll solve the halting problem and then I'll apply that to the problem I'm actually facing" or similar foundational errors. (I say this as someone who has had a 26 year career in software without holding a CS degree.)
- souprock 7y ago
  If you aren't demanding 100% reliable success, "first, I'll solve the halting problem and then I'll apply that to the problem I'm actually facing" is a viable strategy. My workplace does this, especially with the PhD holders. It's kind of amusing to put aside well-founded fears about the halting problem. If you are clever, it works more often than most people would imagine.