5 ms·

For us, when hiring, we explicitly look for candidates without degrees, or degrees in unrelated fields.. they honestly tend to be better hires. They might not k

For us, when hiring, we explicitly look for candidates without degrees, or degrees in unrelated fields.. they honestly tend to be better hires. They might not know some formal definitions or some algos off the top of their heads, but we've found they're generally more resourceful, better at thinking outside the box, and better at solving complex problems without being handheld. Not to mention that what someone learned in uni a decade ago is probably not applicable anymore whatsoever (in tech). Experience/projects are a way, way better indicator of a potential hire's quality than their formal education.

sys_64738 6y ago
CS degrees are not for churning out new hires for whatever you do. They're for providing a well-rounded Computing education which provides the foundations for the student to build on. Things like classic algorithms and foundation computer architecture are classical things that will always help you. Perhaps you're only hiring coders.
- woodruffw 6y ago
  I'm not the OP, but I have a non-STEM degree and work at a research firm doing DARPA-funded program analysis and compiler research. My experience does not match your claim: I've found no relationship between having a CS degree and having a strong grasp on "foundations" (which foundations?). The strongest indicators that I've found for whether someone can do complex, research-level work are: * Being able (and willing) to try at a problem, even with partial information. Even more critically: being able to subdivide a problem such that they can solve some of it and be forthright about being unable to solve the rest. * Just knowing a lot of productivity- and thought-enhancing stuff. My best interviewees are active and vociferous readers, both of technical and non-technical materials. They almost always end up treating interviews as conversations, ones where both parties end up learning a bit.
908B64B197 6y ago
It's an interesting take. I wonder which market you are hiring in. There might be a selection bias at play here. Someone who was able to create its own degree from scratch and learn enough to pass your interview should adopt the same mindset when tasked with a new problem. You are selecting for folks who require less hand holding. > Not to mention that what someone learned in uni a decade ago is probably not applicable anymore whatsoever (in tech) Not if they had a good degree. Sure the tools change and whatever language/framework too but the fundamentals don't. Applying a a systematic approach to debugging, algorithms, automata and complexity are still very much the same than they were a decade ago.
- parliament32 6y ago
  The bias comes from me. I'm the guy who got a CS degree a decade ago. I can tell you from personal experience, 100% of tech-related course material in uni fell into one of two categories: either "I know this already" or "this will never be applicable in a real-job scenario". And I know I'm just some guy on the internet, but I went to a "good" university, and I'm director-level at a tech company you've heard of. As you say, "applying a a systematic approach to debugging, algorithms, automata" is absolutely essential in what we do, but I'm asserting you're much better at all of the above by researching, discovering, and applying it for yourself rather than memorizing textbook definitions and answering multiple choice questions.
  - 908B64B197 6y ago
    If the degree is memorizing textbooks and mostly multiple choice questions then I question the validity of the degree itself. Going to college forces someone to learn full time, and that's the real value. Sure you can learn on your own part time but it'll take wayy longer than dedicating a few years to the problem. So in that regard, I wouldn't advise a prospective CS student to go do something else and learn software "on the side".
kazinator 6y ago
> Not to mention that what someone learned in uni a decade ago is probably not applicable anymore whatsoever (in tech). Working all night to meet a deadline is applicable. You learn dedication at university. University also teaches you to take a risk on something long term. You pay a bunch of money, and work hard without compensation for a number of years, hoping it will pay off somehow. Your risk not making the grade and dropping out, losing the investment of time and money. I remember one assignment which was implementing floating-point addition in 68K assembly code. (Third year course on machine architectures.) Code was executed on an emulator that could provide cycle counts, and there was a cycle count maximum to meet on certain test cases. Now the permanent lesson here wasn't 68K coding, and that's not what the experience memorable. Rather, the day before the assignment was due, the prof surprisingly dropped the cycle count requirements to significantly lower values. Everyone had to scramble to squeeze their code down, even those who had not procrastinated at all and completed their code days in advance. We learned about dealing with a change in external requirements close to the deadline, and still making it.
- simplyaccont 6y ago
  >Working all night to meet a deadline is applicable. You learn dedication at university. It's not dedication. It's simply a deadline that you must meet or face failure otherwise. >University also teaches you to take a risk on something long term. You pay a bunch of money, and work hard without compensation for a number of years, hoping it will pay off somehow. Your risk not making the grade and dropping out, losing the investment of time and money. Actually university is opposite of taking a risk. It's a safe choice. You get a degree - you get a good job. It's something that always drilled into everybody. Taking a risk is doing something that is opposite of common sense.
  - kazinator 6y ago
    > You get a degree Not a given. A kid just has to walk down that hallway where the yearly grad photos hang, and take note of how less populated they are than just that one lecture he or she just came out of.
wiredfool 6y ago
There are bits here and there that apply. "Someone on this job has to know WTF is going on. And that person needs to be you." It was construction management, but yeah. It applies to Project management as well.

sys_64738 6y ago

CS degrees are not for churning out new hires for whatever you do. They're for providing a well-rounded Computing education which provides the foundations for the student to build on. Things like classic algorithms and foundation computer architecture are classical things that will always help you. Perhaps you're only hiring coders.

woodruffw 6y ago
I'm not the OP, but I have a non-STEM degree and work at a research firm doing DARPA-funded program analysis and compiler research. My experience does not match your claim: I've found no relationship between having a CS degree and having a strong grasp on "foundations" (which foundations?). The strongest indicators that I've found for whether someone can do complex, research-level work are: * Being able (and willing) to try at a problem, even with partial information. Even more critically: being able to subdivide a problem such that they can solve some of it and be forthright about being unable to solve the rest. * Just knowing a lot of productivity- and thought-enhancing stuff. My best interviewees are active and vociferous readers, both of technical and non-technical materials. They almost always end up treating interviews as conversations, ones where both parties end up learning a bit.

908B64B197 6y ago

It's an interesting take. I wonder which market you are hiring in. There might be a selection bias at play here. Someone who was able to create its own degree from scratch and learn enough to pass your interview should adopt the same mindset when tasked with a new problem. You are selecting for folks who require less hand holding. > Not to mention that what someone learned in uni a decade ago is probably not applicable anymore whatsoever (in tech) Not if they had a good degree. Sure the tools change and whatever language/framework too but the fundamentals don't. Applying a a systematic approach to debugging, algorithms, automata and complexity are still very much the same than they were a decade ago.

parliament32 6y ago
The bias comes from me. I'm the guy who got a CS degree a decade ago. I can tell you from personal experience, 100% of tech-related course material in uni fell into one of two categories: either "I know this already" or "this will never be applicable in a real-job scenario". And I know I'm just some guy on the internet, but I went to a "good" university, and I'm director-level at a tech company you've heard of. As you say, "applying a a systematic approach to debugging, algorithms, automata" is absolutely essential in what we do, but I'm asserting you're much better at all of the above by researching, discovering, and applying it for yourself rather than memorizing textbook definitions and answering multiple choice questions.
- 908B64B197 6y ago
  If the degree is memorizing textbooks and mostly multiple choice questions then I question the validity of the degree itself. Going to college forces someone to learn full time, and that's the real value. Sure you can learn on your own part time but it'll take wayy longer than dedicating a few years to the problem. So in that regard, I wouldn't advise a prospective CS student to go do something else and learn software "on the side".

kazinator 6y ago

> Not to mention that what someone learned in uni a decade ago is probably not applicable anymore whatsoever (in tech). Working all night to meet a deadline is applicable. You learn dedication at university. University also teaches you to take a risk on something long term. You pay a bunch of money, and work hard without compensation for a number of years, hoping it will pay off somehow. Your risk not making the grade and dropping out, losing the investment of time and money. I remember one assignment which was implementing floating-point addition in 68K assembly code. (Third year course on machine architectures.) Code was executed on an emulator that could provide cycle counts, and there was a cycle count maximum to meet on certain test cases. Now the permanent lesson here wasn't 68K coding, and that's not what the experience memorable. Rather, the day before the assignment was due, the prof surprisingly dropped the cycle count requirements to significantly lower values. Everyone had to scramble to squeeze their code down, even those who had not procrastinated at all and completed their code days in advance. We learned about dealing with a change in external requirements close to the deadline, and still making it.

simplyaccont 6y ago
>Working all night to meet a deadline is applicable. You learn dedication at university. It's not dedication. It's simply a deadline that you must meet or face failure otherwise. >University also teaches you to take a risk on something long term. You pay a bunch of money, and work hard without compensation for a number of years, hoping it will pay off somehow. Your risk not making the grade and dropping out, losing the investment of time and money. Actually university is opposite of taking a risk. It's a safe choice. You get a degree - you get a good job. It's something that always drilled into everybody. Taking a risk is doing something that is opposite of common sense.
- kazinator 6y ago
  > You get a degree Not a given. A kid just has to walk down that hallway where the yearly grad photos hang, and take note of how less populated they are than just that one lecture he or she just came out of.

wiredfool 6y ago

There are bits here and there that apply. "Someone on this job has to know WTF is going on. And that person needs to be you." It was construction management, but yeah. It applies to Project management as well.