Hacker News Reader

Comment by samatman | Hacker News Reader

The tech industry has a reputation for being hard on older engineers, and I’m sure there is some of that, but I doubt it’s the problem it’s made out to be. Anectdote, I know, but the older engineers I know are doing fine, one gentleman pushing 70 recently came out of retirement to work on Ethereum.

The number of developers has roughly doubled every five years for about thirty years now. Even before you subtract programmers who moved into management or got out entirely, there’s just not going to be that many.

This is more for y’all youngsters who are afraid you’ll be put out to pasture when you turn $age-you-think-is-old. No reason to think it’ll happen, experience is useful, you’ll be fine.

Yes, you’ll find some ageist managers out there, who won’t hire you if you’ve been legally drinking for more than ten years. Good! You dodged a bullet there, such people invariably have other bad habits.

View on HN

This kind of ageism isn't really visible to you unless you're the subject of it. Yes, it's real.

Society in general is pretty hard on older people and it increases with age. Think of the way older women are treated as “grandmas”. They can be shoehorned into a role they may not want to play.

> Think of the way older women are treated as "grandmas". They can be shoehorned into a role they may not want to play.

The (American) Jewish term of endearment for a female baby is bubbeleh, which translates roughly as "tiny future grandmother." On the surface it might look like shoehorning, but I think it's meant to be aspirational more than limiting. There may (or may not) be something similar at play here.

I had a specific response to the stereotypical grandma role. I heard it somewhere too long ago to justly represent. But I think it was that grandmothers can be treated like revered figures but generally not taken seriously when they voice concerns or opinions. I wish I had a clearer memory of this point of view...

For men and women it tends to be equal and opposite. Young men are at their lowest social value when young, they're disposable, expected to destroy their bodies to make a living, etc. As they get older their social value goes up. It's the opposite for women whose social value peaks at 18 and declines over the years with a significant drop in the 30s when fertility wanes.

I suspect the (young, male) workers at the FAANG and the (older, male) workers in this thread would be counterexamples.

They're the exception that "proves" the rule, really. There's a quote from someone that goes along the lines "Young women are generally the most valued people by society. The people least valued by society are young men." That's probably what the previous comment was rephrasing. I experienced this first hand, as a man, after graduating from college at the start of the great recession and not having any luck finding work for years thereafter.

Having said that, the economy has been very different for this cycle than it was when I was fresh out of college. It clearly varies inside and outside SV, and all the other usual caveats.

For example, I find that technical interviews for experienced developers tend to be harder, which makes me infer an "up or out" culture. A dev with 20 years of experience isn't even evaluated for normal individual contributor roles, instead they are chief code wizard or unqualified.

Younger devs, especially college hires, are given more leeway to train on the job and grow into their level of aptitude and interest. A similarly intelligent and driven middle aged dev won't be offered the same chances.

What do you recommend to handle that? I find it difficult to hire senior developers for junior positions with the salary that goes with it.

The biggest obstacle to middle age workers who are shifting career is to be unable or unwilling to take the paycut and downsides that come with a fresh start.

Ironically I would be fine with that but the recruitment agents (and everyone else) kept telling me to aim high.

The higher offer you get, the more the recruiter gets paid. Consider the interests of the advice giver.

It's a strong red flag that you are seriously underpaid if the recruiter is insisting on asking for more.

The recruiter has little interest in you being paid more. He's strongly incentivized to push candidates toward the average of the company.

Going over average is risking the company to bail out, too expensive. Going under average is undermining your credibility and leaving commission on the table.

> I find it difficult to hire senior developers for junior positions with the salary that goes with it.

Make the offer. If they turn it down, let them know they can change their minds later.

I always tell engineers looking for jobs that saying "no" is the employer's job. It goes the other way too. Offer the engineer the associate engineer job (or whatever the title is). Let them counteroffer or decide it's a bad fit.

I find it difficult to handle in practice. That requires to adjust the interview and the rating, to conclude that the guy should join as a junior, instead of being rejected.

But there is no shortage of new graduates and juniors with too little experience. What if the company already stopped hiring juniors because they are too many and there are not enough senior to train them? We have no job to offer at that level, it's not about age.

I recommend it at least once a month. I don't know why copying the candidate info to another req should be difficult compared to the rest of the interview and recruiting process.

If you don't have junior level openings, don't offer the job, clearly.

Then the risk is that they take the job, but only for 6-12 months, and then leave just when they're starting to become productive with the new tech and your environment.

Presumably they would ramp up their pay and responsibilities according to their development.

If they're only worth introductory pay and responsibilities 12 months in, let them go take a pay raise elsewhere.

The attitude that pay only goes up incrementally year to year is fairly infounded. The curve should be steeper six months to three years in, depending on you industry and stack.

Organizations should also optimize their tech for onboarding efficiency. I rarely see that enter the discussion other than broadly: "It's easier to find JS developers."

I think there’s an element of expectation there. The college hire might grow to be a grand wizard in 10 years, and become a massive asset to the company.

It is still possible the older person is going to become a wizard later, but the question is: if they have it in them, why aren’t they a wizard already?

College hires are seen as unscratched lottery tickets, older workers are seen as known variables.

I’m not saying this is fair or clever, but this seems to be a part of it.

> but the question is: if they have it in them, why aren’t they a wizard already?

An aspect of this thinking is that people don't always know how to identify those 'wizards' if they can only evaluate someone against the company's own internal tech/lingo.

Someone very well might be a 'wizard' - demonstrably so with experience on visible projects. But if the interviewing staff can't parse out those skills in to something they grok, there's still a mismatch.

Another way of putting this: I've seen people in companies who were considered 'wizards' because they could do everything 'the company way', even though the 'company way' was demonstrable and horrifically both inefficient and insecure. Boat-rockers were not welcomed, regardless of the potential for positive impact.

I think that's one of the very crucial elements of the ever-so-mythical "cultural fit": namely that most companies hire people that look like the median value of all employees already hired. They often don't seem to look for the best they can find.

>but the question is: if they have it in them, why aren’t they a wizard already?

Not everyone started in college, or started college at eighteen.

> The college hire might grow to be a grand wizard in 10 years, and become a massive asset to the company.

I'm not sure if employers are aware of this, but people staying at companies over 10 years are getting exceedingly rare.

Only because employers don't orient pay and promotion schedules around market rates and experience levels.

If you're ten years at the same company, it's exceptional that you couldn't get an offer for significantly more money. Paying "market rate" is a policy for new hires, not existing devs as much.

> Paying "market rate" is a policy for new hires, not existing devs as much.

And as you pointed out, that's a huge mistake. To this day I cannot understand why people simply don't do meetings where somebody says "okay guys and girls, we have a new need in the company; you 5 here aren't heavily loaded -- or we can relieve one of you of their current duties to work on the new thing -- is somebody up for it?"

I have witnessed such an approach I think twice for 17 years of career. And it worked really well the both times. Re-negotiating salaries is a matter of understanding and realizing that you now bring more value and/or carry more responsibility. For some reason HRs and CEOs absolutely hate raising wages of existing employees. They might hire somebody new at 3x the salary of a current senior engineer (seen that as well, and the new hire had significantly less expertise!) but they will be damned before they allow that current engineer get a higher wage.

It might be that they are afraid that if they raise one wage then everybody will demand the same. But I don't see how this would be different with a new hire. No more than 2 months in they will be sharing details of their arrangement with the other colleagues during the lunch break. Same thing, cat get outs of the bag sooner or later.

That phenomena -- huge resistance to raising wages of existing employees -- remains a mystery to me to this day.

But what are the chances that the college hire who had became the wizard is still in that same company after 10 years? It's ridiculous to hire with that kind of long-term expectations.

Close. But I think it's more along the lines of:

"This kid (i.e., recent grad) reminds me of me. My shit doesn't stink. With my care, I bet his won't either."

"There's no way I wanna be that old guy / gal. I deny the fact that he / she could be my future."

Basically, ego vs fear.

I suspect this confirmation bias contributes to various isms. Racism, sexism, etc. That is, they are not me, and I refuse to see they can be me.

"This kid (i.e., recent grad) reminds me of me."

Ah, "culture fit". Looks like me, dresses like me, educated like me, thinks like me.

> Looks like me

The classic signs of 2 words till racism.

As mentioned by another commenter, it’s not actually easy to tell in an interview if someone is a “wizard”. And in fact if that someone has experience different from yours, you are even less well equipped to make that determination. It’s a sad irony that the person best equipped to contribute something new to the team is often overlooked exactly because the team doesn’t recognize their differing expertise as valuable.

The only dev I've worked with that I'd elevate to the wizard level... interview was a disaster, it was his first out of college, he was shy, answers were short, etc. We hired him anyways to get favor with his family, figured we would churn him out and keep everyone happy.

Best hire ever! The quiet shy kid could code circles around anyone id met prior.

Oddly he hired someone else with same degree from same well respected school, guy was not even one tenth as useful as the first hire. Third hire from same uni was junk too. That's when I really stopped looking at pedigree or even what degrees the person has, it's never proven to be a great indicator to me.

Edit;spelling.

> We hired him anyways to get favor with his family, figured we would churn him out and keep everyone happy.

If you don't mind my asking, which country were/are you located in and what line of business were/are you working in?

At that time, Austria doing Ecommerce and enterprise web dev.

> It is still possible the older person is going to become a wizard later, but the question is: if they have it in them, why aren’t they a wizard already?

The vast majority of coders are not and will never be a "wizard". And that's OK. You can still be valuable to a company if you aren't a recent college graduate or a wizard.

Edit: also, be ready to pay through the nose for any actual wizard you want to hire. They can demand bucketloads of money.

My experience over 25 years suggests that wizards can readily command 1.5-2x and deliver beyond 10x. If you can find and engage them, they're one of the best semi-secret bargains in tech.

But to attract them usually requires a lot more then money. You're project needs to interest them, to start.

Every "wizard" I know makes 500k-1mil, some in all salary. I guess that's 1.5-2x some places.

1.) Average stay in company is 2 years, so whether someone grows wizzard in 10 years is immaterial.

2.) If you are hiring for vague sense that person grows in 10 years, you are extremely likely to be biased for all kind if cultural and demographic signals that have zero with actual performance.

3.) Old are not wizards now, for dame reason why young are not wizards now. Nobody can know everything and wizardly is moving target - half composed of myths and half of smokescreen.

It is not to say that great people don't exist, they absolutely do. But they all have expertise in what they do now and when they change jobs, there is learning curve.

> Nobody can know everything and wizardly is moving target - half composed of myths and half of smokescreen.

These people do exist, though. Just like in any sufficently deep intellectual field, there are people who are very gifted and have insane focus. Compounded over years, they grow to be real beasts, just listen to any talk by John Carmack or Jonathan Blow. The bad news for the employees is that they tend to start their own companies, as typical firms don't have appropriate spots for them (as someone else mentioned, these people provide 10x value but are typically only paid 1.5x-2x).

The other thing is, those 'wizards' typically have deep knowledge, not broad. Would most of either of the two people you mentioned really be all that wizardly at an average web startup?

Sure, a games company or similar, but would Carmack be a 'rockstar' on something like a node.js stack? I'm sure he could pick the tech up quickly, but most of his skills lie elsewhere. And yet there is the expectation in tech that truly experienced people will be good at everything, just look at the post about 'things I don't know' from yesterday to see examples of this expectation.

https://news.ycombinator.com/item?id=18780065

I mostly agree. Talented people can often demonstrate that talent across a wide range of situations. But it's naive to expect truly world-class expertise developed over decades to be readily transferable to other areas. It's easy to say people should keep their skills up to date. But if your compensation and reputation is based on a deep understanding of a narrow area that took years of experience to create, it's easier said than done to just learn something new and expect to be compensated similarly.

Sure. And they are great in their area of expertise and able to learn new areas too. When they are changing to slightly new area, they look less like wizards unless given benefit of doubt.

But also, they tend to be less of talkers and presenters which is another separate skill. As in, they are rarely public face.

I'm going to assert without evidence that no one in this industry is thinking 3 years out, much less 10. Heck, the chance someone you hire today is going to be working for your company in 10 years approximates 0.

Why ever hire people with no experience then?

1. You need somebody.

2. They're cheap.

3. They went to the same school/non-school/frat/boot camp you did.

4. They look like they'll be fun to hang out with.

5. You also have no experience.

who stays at a company for 10 years?

Unfortunately I might fall in this category. I am a Sr level manager/influencer at a Dow 30. Ex-programmer. Been there 7 years. I just got accepted at kellogg emba. So unless I switch jobs after 1 year I will be this guy.

One of the reason i got stuck is that 3 years into my job; my ex-VPs, who are both CTOs now, asked me to build cloud based eCommerce platform for this top 5 online retailer.

where i was I charge of digital transformation for the entire online space except the infrastructure.

This was 3 year journey. That put it into 6+ year mark.

It’s less common than it used to be but I was at one job 13 years, another 8, coming up on 9 at my current one.

I'm 54 and have been working in Silicon Valley/SF since '96.

Ageism: One way to combat it is to have a good network. Of the 7 jobs I've had in the last 20 years, 6 of them were from personal reference (schoolmate, co-worker, friend). You don't need to be a badass, just dependable.

Work around your limitations: I'm not as sharp as I was in my forties, and this is especially apparent when I pair-program with someone half my age. Younger developers have an astounding short-term memory (or, as my brother likes to phrase it, "they have a much larger register set"). As a simple example, we have 3 environments at work, and at any given moment they're being used for different purposes. I write their purposes down on a scrap of paper to keep it straight, but my twenty-six-year-old pair keeps it in his head.

Stay fresh: I know a 69-year-old developer/contractor say to me, "2000 was pretty good but after that the work dried up — I guess they didn't need COBOL programmers anymore". He had never bothered to stay current.

Make no mistake: picking up new skills can be hard. I remember in '86 when a developer in his 50's turned to me and said, "Brian, I'm tired of learning new things." He really liked to bartend though, so I hope that career worked out for him.

And some developers aren't interested in working anymore (I think this can strike at any age). I remember five years ago pair-programming with a developer my age who wanted to reminisce ("Remember when 640kB was a lot of memory?") and show pictures of his grandkids. I felt bad when they let him go, but it was hard getting work done while I was pairing with him.

> Stay fresh: I know a 69-year-old developer/contractor say to me, "2000 was pretty good but after that the work dried up — I guess they didn't need COBOL programmers anymore". He had never bothered to stay current.

This is one that a lot of people have trouble with. One of the best engineers I ever worked with was in his 60s at the time, and seems to have stopped learning new organizational systems sometime in the 1980s. Agile development processes were just newfangled bullshit to him.

He was hot shit at anything that required fine attention to detail, and implemented a distributed task system after I gave him a verbal thumbnail sketch. Correctly. The first time.

But I could see that his refusal to learn new things was interfering with his career all the same.

I specialize in modern Python and to some extent JavaScript. Necessary but not nearly sufficient to get hired these days.

what if they actually are newfangled bullshit? (a mere proposition, as fads fade. I merely ask you to entertain this proposal)

It sounds like another entity might be able to use his brain a bit more thoroughly.

> what if they actually are newfangled bullshit? (a mere proposition, as fads fade. I merely ask you to entertain this proposal)

That might be true and both you and he correct in every single way! But in this particular context it wouldn't have mattered one whit - he and I were working on a contracted project where the counterparty had mandated agile and it was written into the contract. It didn't matter if he liked or believed in agile methodologies. What mattered was whether he would learn them enough to be an effective team lead. He refused.

Quite often, new things are actually newfangled and universally less useful than extant tools. In such cases, I expect the wiser heads who can spot that to be able to clearly and compellingly make the case for why this is true in every scenario where one might consider using the new and inferior tool. He did not and could not.

Again, you're completely right. Lots of things are just value-free newfangled bullshit. There just might be a subtle distinction to be made between an old hand pronouncing something as such and it being an incontrovertible fact.

Funny note: the original paper for Waterfall had many agile concepts built into it, if you read more than the first page.

I think it really depends on lots of factors. My father joined a startup in the late 90s, after 5-6 yrs the company failed. At 60 he was unable to get a new job. I'm guessing companies assumed he'd leave in 3-5 years and didn't want to invest the time.

I also know from running my own company with partners back in the mid 90s, we were all early 30s except one partner who was 52ish. When getting health insurance for the company his insurance was much more expensive than anyone else's. Maybe that only happens in places like the USA but it registered as yet another reason a company might not want to hire an older person.

> I'm guessing companies assumed he'd leave in 3-5 years and didn't want to invest the time.

3-5 years is a pretty long stretch in this industry. The longest I’ve ever held a job is just barely 5 years. Most tech employers— at least the smart ones— know the average bounce rate in the industry is high. Older employees tend to value stability, for various reasons. So I don’t buy your argument.

If there’s a reluctance to hire older devs, I suspect it’s because they have higher salaries and maybe because they are (inaccurately) perceived as being unlikely to know or want to learn new tech.

The neighbor I mention in my other post told me he'd work for $30K if they offered insurance and he'd stay for at least five years. As a software engineer, I know he can run code rings around anyone.

Yes, the salary requirements thing isn't very convincing to me. At the end of the day a low salary is better than no salary. Also people approaching their sixties will have most likely paid off their mortgage and their children will have left home.

>3-5 years is a pretty long stretch in this industry.

That's a rather SV-centric view. Otherwise 5-10 years (or longer) is quite normal, including in tech. At many companies, a bunch of 3 year stints for a job candidate is something of a red flag.

I’ve never worked in SV. Only small enterprises, Microsoft, a handful of local startups, and one SV company. The SV company had above average retention compared to most of the non SV places I’ve worked. The 3 year range has been pretty much the norm in my experience (20 years or so). 5 years has been on the high end in every company I’ve worked for. It is anecdotal, but it is what I’ve seen consistently across markets both large and small.

> That's a rather SV-centric view.

Trust me, it isn't. It's valid for every organization I've ever been in or consulted for (15+). I am in Eastern Europe and spent 9-10 years in offices, and the other 7 after that it has been all remote. I've seen a lot of companies for both consulting and normal 5-to-9 work.

While the reasons why most people don't stick around above 2-3 years are many and varied (and some are frankly quite stupid) to me it does seem to be the norm.

But I'll readily admit we all live in our own bubbles and confirmation biases. So I'm just sharing an additional anecdote here.

Fair enough. I’ve worked mostly in the US Northeast and while I’ve worked in companies across a wide range of sizes my observations are probably flavored by a lot of big firms who have been partners and clients. I would have said 5-10 years was more the norm among people I work with but then I probably tend to disproportionately work with people who are more senior in their organizations.

What's funny is that many young people in the industry now would never think of staying at one place for 3-5 years cause they don't want to invest the time.

A neighbor of mine is in his 60s and lost his job the same way. He can run rings around anyone I know but he can't find a job or even get an interview.

Bjarne Stroustrup, the C++ creator, wrote a blog post about going for a job interview and being told he "didn't fit the culture" and he knew what that meant.

Not being able to have an interview is a red flag. Companies can't tell the age from a resume.

If they can, take off the date of birth, remove the year on your degrees and don't list experiences beyond 20 years.

>Companies can't tell the age from a resume.

Companies can't add 18 + length of schooling + length of career ?

You shouldn’t list your entire career on your resume, just the past 10-15 years. Also, assuming “18 + length of schooling” is bad too. You’d be way off on my age then too (I graduated HS several years early).

> Also, assuming “18 + length of schooling” is bad too.

That does not mean that it doesn't happen. I mean, we already established that we're dealing with short-sighted people here.

> That does not mean that it doesn't happen.

Never said it doesn’t happen, I know it does. I simply stated it’s a very flawed approach and gave one example of why.

18 + 4 + 10 = 32

You actually have 18 years of experience, making you 40 years of age, and you're listing only the most recent ~10 years. They have no practical means to accurately figure out your age from that resume. The best it can do is slightly help you avoid the first layer of age-based screening.

I've done this on mine, and I'm pretty sure when companies see this, it's a signal for an older person.

Then you show up for the interview and people still can't tell your age? Isn't that a waste of time?

I've been given this advice and I'm skeptical. You can only mask things to a certain degree, it can make you look like you're hiding something (which you are), and the hiring manager will figure it out soon enough in any case. If they really want to age discriminate you're probably better off finding out sooner than later anyway. In any case, later career jobs tend to come from personal networks anyway. My resume was just a formality for any interview I've had in the last 20 years.

It's not hiding anything. A resume is a two pages flyer to advertise your services. It doesn't pretend to accurately represent you as a person or 40 years of accumulated experience.

It's really bad if a resume never pass the first screening, it literally can't get a job. It's not reasonable to blame age for that, interviewers cannot guess the age from just a resume (and if they can they shouldn't let be).

Yes, it’s a flyer but it’s a flyer that many expect to include specific information in a specific format. You are free to ignore those expectations of course but employers are also free to draw various conclusions about apparent employment gaps, incomplete work history, etc. even if those conclusions aren’t warranted.

can and do. your premise is faulty

> Bjarne Stroustrup

I actually spoke to him at a tech event, he was super nice and I totally don't doubt the 'culture fit'

Any links to the Bjarne story? I can't find it.

Wow, do you have a link to the blog post? Bjarne, if you read his C++ books, is a great technical communicator and author, and his contribution to computer science speaks for itself.

Shhhh, you’re busting the narrative. But seriously, I’ve worked with some amazing older engineers. I’ve always had great relationships with them, and I’ve learned so much from those kinds of people. In my experience many older engineers are much more chilled out and willing to give their time to help others.

Flat out, they have experiences. Sharing those matter. Maybe the other person can avoid more bad experiences.

I love working where there is a mix of young and old, where they get along.

Same here. I must say, in my work experience, I had problems only with engineers as young as me (or around my age) who had a big ego, a very bad personality or the I-am-very-smart syndrome.

Instead usually the "old" engineers who stick in a tech position were always extremely humble even if having a ton of domain and company-specific knowledge with the willingness to share all of it.

Yeah. I’m very lucky to have worked with quite a few of them. I’m definitely better off for it. I’d try to spend as much time as I could talking with them.

Even now that I’ve been working in senior positions for years, I’m still just as grateful that I get to work with people who know a lot more about stuff than I do.

To be honest, you're probably on to something. Outside the tech hubs, older workers are common (and if you don't want to work those 80 hour weeks, you probably want to be outside the tech hubs, although you likely also won't strike it rich).

And if (career, area) is 80% under 30, any given company is mostly under 30, and naturally they're going to want to hire under-30s.

> one gentleman pushing 70 recently came out of retirement to work on Ethereum.

I think we are talking about 45 yr old with 2 kids and a mortgage.

Reply on news.ycombinator.com