Beware tech career advice from old heads – Jacob Kaplan-Moss | Hacker News Reader

Beware tech career advice from old heads – Jacob Kaplan-Moss

jacobian.org

35 pointsraybb46 comments

This is a big problem with younger devs right now, too.

There are a lot of young developers who got new jobs in the 2021-2022 job market who don’t have any other frame of reference. They preach their experience to even younger devs and college grads as if it was normal, without understanding how unusually good the job market was at that time.

Browse any of the advice forums and you’ll find people confidently telling others to quit jobs they don’t like because it’s really easy to find a new job. Then you ask for details and they reveal that they did this in 2021 and it was fine.

I can’t count how many times I’ve had to tell young developers to ignore bad advice from people with 5 YOE who have never experienced a difficult job market.

> The tech industry of 15 or 20 years ago was, again, dramatically different from tech today. I used to joke that if you knew which was was up on a keyboard, you could get a job in tech.

Nope, guys that started 20 years ago came into very brutal tech job markets. Dotcom bust / 911 / with the cherry on top, the GFC a few years later. Very worth listening to.

And the 80s, and the 70s…

The modern day tech career required two things to exist. A PC in every persons home/pocket and a connection to the internet.

Both of those things came together in the mid/late 90's and gave birth the the modern day tech career. Since then the modern day tech career has changed very little if at all.

Anything before that was not even comparable. Mostly academic researchers and labs.

A bit earlier I think. Most businesses had PCs and LANs by the late 80s. I specialized in Netware and PC repair in the early 90s. Some graphics/desktop publishing on the side for fun.

The difference between the tech world in the late 80's (computers and software are common and important) and the rise of the smartphone/web eras is so massive as to be fundamentally different.

Yes, we had cars that existed in the 20's and 30s, but listening to those "oldheads" about what worked for them at that time if you were in the post-WW2 auto boom in the US wouldn't make sense. See also: aerospace pre/post jet propulsion.

Nah, my day is about the same. Sitting in front of a computer hacking and making better than average pay.

My time outside of work has changed substantially however.

So your point is that youngsters are the only ones that understand “modern” software? Or economic crisis? Or both?

Stereotyping is fine as long as it’s understood to be fundamentally wrong and inefficient, and when applied to people cruel.

What are you on about? Where did I say that?

My point was: Saying "this is what worked for me finding a job in the 80's/90's" isn't necessarily great advice, because it's not the 80's or 90's. If an oldhead says "this is what worked for me last year" sounds like he's giving useful advice.

Put succinctly: Oldhead advice is great when it's relevant, but oldheads need to understand that not all their war stories are useful. Feels like an extremely reasonable take, but strawman me some more if that's what gets you off.

Signed- An oldhead!

I still remember the NetWare/PC guy coming to our house as a kid when I bricked one of my families early pc's. He said something like, "if you keep this up you will have my job some day."

We still didn't have the internet in our houses then, though.

Having decades of experience doesn’t necessarily blind one to the reality of less tenured folks. The reality is that there are externally observant and empathetic people at all ages who see and internalize the cause and effect patterns around them, and those that aren’t and don’t. The latter are for some reason more vocal than the former lately.

I remember being told to 'specialize in something' when I first started. Looking back I think this is terrible advice.

I think "specialize in something" is terrible advice for a junior but great advice for people with 10+ years experience.

If you're an ambitious junior I do think it's best to job hop as much as you can within reason, like every 6-18 months. Partly due to salary increases, yes, but even more importantly, you learn from having experiences at different companies. No two companies are alike, and working your whole career at one company, even if it's an ostensibly good one i.e. FAANG, limits your exposure. I've seen lifelong FAANG coders who tried to join a smaller startup/scale-up and fall flat because they couldn't adjust. Obviously, job hopping is much harder in the current market, but places are hiring, so it's still worth trying.

For seniors though, eventually the value you provide as "just" a senior developer caps out. A software developer with 10 years experience provides a lot of value over a developer of 1-2 years of experience that justifies a significant difference in pay, but does a 20 YoE dev provide a lot more than 10 YoE? In a general "senior developer" role, I don't think so. For specialized roles, though, you can have a lot more influence, provide a lot more value, and yes, command a better salary. Unless you move into tech management, specialization is how you can keep progressing once you've acquired the requisite technical chops.

oof 6-18 months as a hopper? It are really shitty jobs in my experience if that's the max you can stay somewhere before not learning new stuff.

I'd say that just leave once you are too settled and are just performing tricks. That can be 6 months as you say but also 5 years. Also, never be the smartest guy in the room.

context is everything… I compare generalist vs. “specialize in something” similar to using diversification while investing (index funds) vs. stacking your chips into few things.

no WEALTHY person is all that diversified (e.g. warren buffet) but of you course there are plenty RICH people who are. specialization is high risk, high reward (e.g. have a dear friend making insane money as cobol coder for decades now who couldn’t write a hello world program in any other language) while generalist is safer route but won’t get you wealthy, just rich (which isn’t too bad :) )

You think? I only see jobs for people who are specialized in something

> I remember being told to 'specialize in something' when I first started. Looking back I think this is terrible advice.

You just have to specialize in something that will be relevant for 20 years.

Worst case scenario, at least you'll learn about survivorship bias. :)

You can shorten it and say, beware of any advice at all.

Advice is someone trying to crystallize lessons they learned from doing something that worked for them in their circumstances.

Understand their context before deciding what weight to give their advice.

Trust but verify

I'm guilty of giving this but the old head advice that falls short is "use your network". Easy to say when you have one having been in industry for 20+ years. A bit harder to do when that number is zero.

If you're coming straight out of University, there are likely several dozen people who have a reasonable idea of your capabilities -- your classmates and your professors. That's your network.

And the value of that network is often very low in finding a job after university. Your peers are unlikely to be in a position to offer you a job, and your professor may have a connection but realistically this is unavailable to hundreds of grads every semester. When you’re fresh in the career “use your network” feels like the same kinds of job postings that say “junior engineer - must have 3 years professional development experience”.

Also hard to do when you're not a networking person. I'm a programmer for a reason, and part of that reason is that I'm a bit of a misanthrope. A friendly misanthrope, but a misanthrope, nonetheless.

Try keeping in touch by email. I send out a message every new years or two.

> Easy to say when you have one having been in industry for 20+ years

More like 10+. That is the network sweet spot. At 20+ your network is starting to gray out and mostly useless unless you have someone who has risen up to VP ranks who can influence some downstream director to move your resume to the top. At 30+ you are some younger person’s “network”. At 40 years (where I currently am) your “network” is frankly retired, dead, or just trying to squeeze out those last few years in a no stress role to keep their medical insurance and can be no help to you.

YMMV. At 41, I’ve been in tech for about 25 years, and my network is still hugely beneficial to me. They help me keep an eye on new trends, the overall job market, and potential new positions. I focus on keeping in touch with a lot of people, though, even if it’s not constant contact.

For the new grads, it should be “build your network”. Learning skills for promoting oneself is imperative in the early career stage. Start a blog. Connect with a community. Find a meetup or user group. It pays off long term. the unfortunate part is that it doesn’t necessarily lead to a job immediately.

If you’re starting in tech right now, and can go back to live with parents after graduation, then do that while building up your presence. Just don’t take your foot off the gas until you have that solid network. The stigma about moving back home doesn’t make sense to me, unless all you do is play WOW (or whatever people are wasting time on these days).

If that doesn’t work, find some roommates. Be willing to slum it a bit while establishing yourself. It might not be glamorous like Instagram or Tiktok says it should be, but the flexibility you have while keeping costs down opens more opportunities. You can take more risks if you don’t have to worry so much about making rent or paying bills.

I'm hesitant to consider anything said by anyone who uses the term old heads seriously.

Kind of a silly piece. I don't think young people take advice verbatim, obviously it is going to need some adjustment to their current situation.

And, if you read "The Mythical Man-Month," you'll be surprised how little has changed.

Technology changed a lot in the last 20 years. And don't take advice from someone that have been only working at the same place and same technology all that time.

But what about the old folks that have been changing and adapting to the new technologies, work conditions and reality all along the path?

Ok. So why? What's changed?

According to the article: "The tech industry of 15 or 20 years ago was, again, *dramatically* different from tech today."

"Dramatically!" Therefore, old career tech advice doesn't apply anymore. Gotcha. Thanks, Captain Obvious? This blog doesn't really say anything is my point...

The technology, the types of software being developed, what the major players (FAANG, etc) are focusing on, the rise of AI, the wealth of information available on the internet, the hardware, politics, etc. etc.

Interest rates.

To put a finer point on this that others may miss - when interest rates are low as they had been from 2009-2021 “money is cheap.” When money is cheap, companies can spend more of it on growth and when companies are in growth mode they spend more on staffing, and especially dev teams.

So, naturally during that time period the job market was very hot and job hoping was easy. Now that interest rates are higher, money isn’t as cheap, which means companies can’t afford to spend as much on growth, which means that the job market is much cooler.

To elaborate on this, higher interest rates mean that money in the future is less valuable than money today, which means that investments in software etc have a higher bar to pass to be worthwhile doing.

Newer fangs are sharper fangs - don't gum your career options to death.

Closing in on 50 years experience, under 60, USDA Grade A genius and son of one who is closing in on 100 with a bigarse certificate above his bed signed by a US Secretary of Defense..

..Sadly, the man with a law firm of a last name is probably from the era of tech professionals that began to emerge after the (dot-com-crash) nuclear winter ended, when most all over age 40 had been permanently purged from the tech industry for financial reasons. (People over 40 typically cannot survive for 5-6 years without earning an income and have to transition to other occupations in support families/mortgages/etc.) He most likely didn't come up with any grey-beards mentoring him at all. I can't even imagine having education, talent but the wisdom of random Google searches and Reddit posts. You can only go so far in life without knowing the "why" of everything! So, with all due respect, likely coming up without the mentorship trade professionals with even 20 years of experience means he doesn't know what he doesn't know about whether or not to take advice from "old heads." Case closed. (So depressing, how about a near-endless stream of beautiful women https://www.youtube.com/live/gqOeWBxN4_Q)

PS: First new grey-beards since 2000 (outside of Defense/Big Tech, Europe and Russia) should be around soon. A legit example of the handful left from before 2000 are like https://www.youtube.com/@DavesGarage Want some actual "career advice"? Figure out why Dave knows more about DeepSeek R1 than everyone YOUR age? (Hint: None of those people are going to make it in this field to Dave's age.)

its also an interesting prisoner’s dilemma:

telling people what works for me, crowds out what works for me

I see lots of takes on this thread alone, tempting to comment on them but it messes with the money

this basically related to any career advice, no matter if it is tech or not. Too generic an advice

Tying an onion to your belt, like in the old days, is a sure way to a new career.

"My advice is to not take my advice. But seriously, take my advice!"

It's a toughie. I like the Marshall McLuhan quote “We look at the present through a rear view mirror. We march backwards into the future” [1] [2] but I also bemoan what I call the "ahistoric turn", the tendency I perceive that people today aren't interested in the past at all. I can point to many old obscure books that were quite prophetic [3] [4] and lines of research such as neuro-symbolics in the 1990s that went from science fiction to obvious. In the late 1960s, before the CMOS transistor was established as the universal computing element there was a lot of interest in neuromorphic computing and speculation about the "intelligence explosion" that was fresh, not the dogmatic and narrow-minded dogma of the Yudkowsky cult.

Lines of argument that have long gone out of favor are likely to come back [5] as much as some concerns will be reversed [6] and one book that spoke to the zeitgeist of the 1970s is accepted in Japan as an interpretation of the Tokugawa period [7]

One of the ways to deal with rapid social change is to extend your reach to the widest range of cultures and history.

(Funny I was talking the other day to a coworker from a previous job who started programming on 370 mainframes in the 1970s and was telling me how he built a test harness for fly-by-wire control systems for military aircraft based on a cluster of 80386 PCs in the later 1980s.)

[1] https://www.goodreads.com/quotes/618118-we-look-at-the-prese...

[2] https://www.amazon.com/Forward-Through-Rearview-Mirror-Refle...

[3] https://en.wikipedia.org/wiki/The_Image:_A_Guide_to_Pseudo-e...

[4] https://www.amazon.com/Information-Machines-Their-Impact-Med...

[5] https://en.wikipedia.org/wiki/The_Limits_to_Growth

[6] https://en.wikipedia.org/wiki/The_Population_Bomb

[7] https://en.wikipedia.org/wiki/The_Culture_of_Narcissism

Thank you for the list, seems interesting indeed.

Reply on news.ycombinator.com