Ask HN: For older devs, do you feel like you have missed your prime time?

[abdnafees]

I am 27 years old. Did my bachelor's in CS in 2018. Worked in a non tech role for SAAS product for 3.5 years. Moved back to engineering last year. I am now working on a PHP backend and React on the front.

I go through various forums and find that people DID A LOT of cool stuff while I was busy playing with Arduinos and OpenCV. I did learn about the kernel a bit but that's about it.The amount of development that I see people have gone through in the the last 5 years is insane. Sure, there is a lot of sugar coating but I still feel the amount of compute and resources that are available now might mark the point of no return. I hope I am making sense here.

[codpiece]

At around 50, I felt washed up; in a dying industry and a failing company owned by private equity. Job was bleak and thankless, I was old and unemployable. If I lost my job, I was hopeless.

A headhunter called me out of the blue. We talked for 20 minutes, he had one -one- option for me that was a perfect fit, and I started working at a boutique consulting agency.

Fast forward 5 years; I'm leading a group of awesome consultants at an even cooler, massive consultancy. I'm doing work for clients I never imagined working with, and learning emerging technologies. Work is fun, clients love us, my team are brilliant and funny, I completely reinvented my career, and I'm making triple what I was a few years ago. Yes, there is a lot of grind, and I don't create like I used to. I don't want to give you the wrong impression, but it's AWESOME nonetheless.

Sure, I did super-cool stuff in 1994, and again in 2004. In the 80's too, if I think about it. I miss those days. Maybe I should have peaked then, but I'm too stupid to stop surfing whatever wave is in front of me.

Opportunities abound for anyone curious and willing to take risk. Age is -not- a construct (aches and pains are true), but as you age you acquire wisdom, and it's surprising that the stupid stuff I say to smart younger people who take it as profound. I'm actually helping people. I never dreamed that 5 years ago.

Oh, hey. Tech is too big to follow. Find your interests and pursue them, leave the rest. I love generative art and followed it for a few years, but it's only recently that it's more accessible. I'm one of the leaders in a 10K+ person org. At 55. In addition to my day job. Just have fun. It all works out in the end.

Life is a jungle. Swing from the vines.

[BrianHutch]

As a dev/architect in my 50s who has been on the sidelines for a few years, thank you for sharing. I'm on a long sabbatical of sorts and I appreciate hearing that getting back into the game is still possible.

[codpiece]

Hell yes! You have seen a lot, and that's the most valuable contribution you can provide. Newer career people get caught up in the drama and can't find a good way out. You've seen this play out hundreds of times. You can provide the calming perspective.

Ping me if you want to chat.

[abdnafees]

Thank you so very much for sharing story. This is so heartwarming. I will surely keep all of your advice in mind. Thank you again.

[codpiece]

P.S.: My wife's most important consultant is in his 80s. He paints, he consults, he wrote the seminal book on his industry in his 60s.

[ilaksh]

27 is not an "older dev". FFS.

[specproc]

I get the sentiment, but more kindly..

I only really got going in the last few years after a long non-tech career. I wouldn't count myself as an older dev and I'm ~40. I play with all sorts of cool stuff all the time, was into huggingface before everything blew up.

You'll be fine. 27 is a great age to enjoy all sorts of things. No one looks back on their life and thinks how they wished they'd worked more.

Fuck the grind. See your buddies. Get laid. Travel. Give your family some love.

[abdnafees]

Yeah, it's not about the age but more about the constant nagging feeling that I get. Anyways, I appreciate the good words.

[cultofmetatron]

> 27 is not an "older dev". FFS.

yea seriously... I'm 38 and I feel like I'm just getting started.

[haxiomic]

Absolutely not, in fact now is the era of the enthusiastic newcomers of any age – it's never been easier to learn something quickly; lead with your ambitions and figure out what you need to create them!

[tizio13]

I can’t second this advice enough. It’s never been this easy to have an idea and also resources to execute on your thoughts.

I have an older(60s) coworker that I’ve been showing the pros and cons of gpt. She’s gained so much confidence from it. She uses it all the time to figure out excel formulas and help with writing emails. She’s bilingual, so it really helps her figure the grammar of her thoughts.

[abdnafees]

I couldn't agree more but there is so much stuff to learn about. Also, learning for the sake of learning is great but I do wish to be paid for my ability to learn at a rapid pace which I am struggling with at the moment.

[zerohp]

Completely agree. I didn't even start my bachelor's degree until 36 years old.

[j0hnyl]

If you're 27 and think you're an older dev, you are lacking some serious perspective.

[knightofmars]

Prime has no other meaning than your expectations of success and when you've achieved them. Getting to know yourself and being comfortable with what it is that you want to do is the most important thing you can do in your life. Worrying about your "prime", sounds more like buying into a hype cycle than anything.

On a personal note, every step of my career has been forward and also unexpected. What I envisioned myself doing and what I ended up doing were two different things. And yet I'm happy and successful. Age 41.

[abdnafees]

Hmm, yes, this is helpful. Thank you.

[Tcepsa]

At 27 I had not accomplished particularly much. I had one solid accomplishment under my belt from my first two years out of college and then bounced around for the next couple of years. I had just accepted a new position at a large, prestigious organization that did a wide range of work... and proceeded to spend the next four years or so muddling along. Doing fine work, but still not really a great fit. Got a Master's in CS during that time, learned some things about graphics programming and systems design and broadened my toolbox a bit... And then in 2012 I got an opportunity to work on a new project, in a different language, outside my group. It was a better fit, still not great but I got to meet new folks and start learning (finally, in my early 30s) to reach beyond the role I'd been given and develop a good reputation outside my group. Shortly after that I switched groups to the one that new project was primarily in. Within a year or two of that I was tapped for a prominent role on something of a long-shot project, and we were very successful, leading to expansion of the effort to other equipment and substantial additional funding. At roughly 34yo I was invited to be the Assistant Project Manager on both efforts to help share knowledge, tooling, and ensure they were leveraging common functionality effectively. This went great for a couple years, at which point I was asked to head up the UI team of another ambitious project that was just getting started. That also ended up being extremely successful and is still running today, but I have determined that that degree of management is too stressful and not very satisfying for me. I have stepped back from formal management positions and have pivoted to technical leadership and contributions.

Now, in my early 40s, I feel that I am finally starting to hit my stride!

[abdnafees]

Damn, that's quite a journey. So happy to read through it. I think I wouldn't say that I haven't accomplished much because if you go through my LinkedIn and know me personally you would say I am in the top 10% in my country but (very sorry but there is but), it's all spread out. It took me years to consolidate part of that knowledge and build a portfolio that includes one of Big 3 consulting firms and I missed out on building cool stuff in my favorite programming languages and tools. Missed out learning from cool courses and much more. I shouldn't do this with myself because of clinical depression but I can't help it.

[BaculumMeumEst]

i think building relationships, finding a partner, and working towards starting a family are a much better focus than “prime development time” in your 20s

[syntheweave]

Prime time is a relative construct - it's a kind of realization of the rat race that assumes that the world is moving in a directed sense "towards" a goal and you are left behind if you don't stay directly, vigorously competitive and master the very latest thing, as if school never ended and you are just proceeding to grade 13, 14, 15...

It can be somewhat true societally speaking, but also insensible as a creature that is part of society. Computers are still basically doing the things everyone predicted they would back in the 70's. The tooling is nicer, but the paradigms of how they should be programmed have crawled forward very slowly relative to the research of that time. So what, exactly, has actually changed? Why do we not have to learn for 50 years to have jobs in software?

Most of what changes in industry is just a "meta strategy", like when you leave an online multiplayer game for a bit and come back and see that everyone is using different builds. Train the muscle memory for that build and suddenly you are competitive again, inside of a year or two. Like when a prize fighter comes out of retirement for a big matchup.

Getting ahead of that and reaching a higher grade of achievement is all in doing the minimum of meta churn and instead finding the thing that differentiates if you go deeper with it than anyone else and solve hairier issues - being the artist who works with unusual media or processes.

And if you're feeling like you aren't working on the right things, then the answer is not desperation, it's to engage with some philosophy and find more logical coherence in how you see yourself, your identity, goals and how you pursue them. What school prepared you for was to know a few things about computers, not to "be a programmer". Being a programmer is ongoing maintenance, a thing of doing enough rote work to still be inside the meta. That's all.

[abdnafees]

So finding a niche. Got it. This does make a lot of sense. Thank you.

[peruvian]

The only regret I have is I spent ages 21-24 thinking I was too cool for "boring" tech like Django, Rails, REST, etc. I was definitely hyped on a lot of the cool tech back in 2013-17.

This did not affect my career that much and I am glad I got exposure to obscure concepts but feels like I had to learn the basics a little late.

[abdnafees]

I have the same feeling. Who knew I would be working on PHP in 2023. And surprisingly it's so much better than JS at the backend at least.

[sethammons]

I didn't really start my developer career until 27. I've achieved a lot since then and I've learned more than I could have imagined and worked on great products with great people. 27 is very young yet. Do cool things and in five years the now-you would envy the now+5-you. To do cool things, work at an organization that is growing.

[greatpostman]

The best thing you can do is work with exceptional developers on complex code bases. You will learn way more than you can alone.

[mgl]

Absolutely not - there is no point of no return for curious guys like you and you will find your self-learning experience that you gained invaluable at some point in the future for sure.

Keep coding!

[abdnafees]

Thank you so much!

[ktbwrestler]

At 28, I feel the same way for slowly meandering around Cybersecurity engineering --> AppSec --> Software Engineering...

I feel like I am a master of none and it really fuels the impostor syndrome.

I think some may be missing the point here that your brain sort of starts to go downhill at this point in terms of fluid intelligence. I feel your pain and hope that we have somewhere to go.

[ktbwrestler]

To follow up in a less morose way... is there anything right now that particularly gets you excited? Are you enjoying any side projects? I think that can really help push people in our position.

[knightofmars]

This surprises me. Having a security background in software engineering is worth an exceptional amount to the right parties.

[thevagrant]

I used to feel like you. At 27 you have a lot of time, keep moving forward.

Industry moves in many directions and some of those roads come to an end.

Don't worry too much what others have done and keep progressing your knowledge with the road you have taken.

Don't be afraid to change path or direction if you feel it is what benefits you, and don't regret.

[giantg2]

I'm in my 30s. I have passed my prime in many areas of life - dev, physical health, etc. I'm working a job paying near the national median in a moderately HCOL with no potential for advancement. That's just how it goes.

[d13]

I’m 52 and in my prime now. I only started my professional coding career at 27.

[unmole]

> I hope I am making sense here

Sorry, no.

[mikewarot]

I'm 59... (I got long covid, my brain is fuzzy, and I can't work more than a few minutes at a time, all things considered, I'd going ok, not great), On the other hand, I feel that the world has taken several wrong turns. I'm interested in correcting course, but feel my ability to help that happen is almost nil. Here's the chain of events, as I see them.

1960s - the military realizes that a single computer can not handle data from different levels of classification. (This was related to planning classified flight operations during the Viet Nam conflict, the flights themselves had to avoid enemy SAM sites (the knowledge of which was Top Secret, even more secret than the flights)), etc... and those were different levels of classification). Research to solve this problem was done, and progress was underway to build this into Multics... when Unix took off, and distracted everyone. There have been some niche secure systems available, but widespread knowledge of them didn't happen. Security of that level wasn't seen as necessary, and eventually was seen as impossible anyway. Note that the solution to general purpose secure computing was found, and proven to work, decades ago!

1970s - general purpose personal computing came along, again without security in mind. BBSs arose, along with UUCP, FidoNet, etc. in the public sphere.... ARPAnet in the Military/Educational area.

1980s - the IBM XT (or clone) with MS-DOS and dual floppy diskettes was the pinnacle of secure general purpose computing. The shareware revolution happened, and most PC users were happy to "buy" $2-3 floppy disks in bulk with various programs from strangers at computer shows, and just try things out.

Why was it secure? A floppy diskette full of data is a course grained "capability". You know (because you insert/remove them, and attach write protect labels) exactly which disks are in the system, can make backups of them easily, and it's effectively impossible to mess up your computer with a bad program.

You also had BBSs from which you could download software to try out. This was peak computer user freedom, even though the machines were slow and the diskettes weren't perfectly reliable. You could just try things, without worry. Nobody has that freedom any more, no matter what OS they run.

The Windows Era - The adoption of hard drives and GUI interfaces brought an end to users having transparent and full knowledge of where and how their data was stored. The need to "install" software transformed what was once a matter of copying a boot floppy into an impossible to replicate system state. Hard drives were expensive, and fixed... you couldn't just copy them freely, like you could with diskettes. This was the first step downhill into the descent.

Still, at this point, there were some great tools introduced at this point. With the Mac, you had Hypercard, on the Windows machines, you could get Visual Basic, or Delphi, and build applications to do CRUD or interact with custom hardware fairly easily. Documentation was included, complete, comprehensive, and amazing.

Then the .NET era happened. This made software slower, there was always a new .NET library to load, and things crashed far more often. While it might have been a good move in preparation for the migration away from the Intel instruction set, that has taken decades, not years, and the framework has been through several incompatible iterations along the way. We lost VB6 and Delphi and Hypercard along the way.

Simultaneously, the Internet was released for commercial use. Eventually, we came to have systems with persistent internet, but operating systems intended for the classroom or small corporate environment. Any thought of security was layered on top, not built in.

Then the web hit, and we shifted from high performance, easy to build and distribute desktop applications to a model where everything is shoved through a stateless protocol through firewalls and proxies to end users on machines they don't fully control, own, or understand. It's a huge mess, and it can't be cleaned up because none of the computers at the edges are secure enough to run random code.

We could fix this... and I've been trying to push that message wherever it seems like the ideas might take hold.... if we abandoned the flawed concept of ambient authority that underlies Windows, MacOS, Linux, etc... and went with one that defaults to no access, such as the ever delayed Hurd, or Genode, then it would at least be possible to get back the ability to run mobile code without risk.

Once that almost impossible task is done, then we can take the code generating tools we built for Windows back in the 1980-90s, like Visual Basic 6, and Delphi, and recast them to generate code to run directly on the phones, tablets, laptops, desktops, etc. The end user can easily manage security with the powerbox facilities that capabilities based OSs provide. (They look just like the file open/save dialogs we're all used to, but then only provide access to those files to the application).

Note that this is NOT the same as "permission management" on your tablet/smartphone.

We could be heading towards a bright secure future, where we all own our own hardware again, and things just work, quickly, without bloat, without virus scanners, the way we want them to...

or not

I think we've got a 0.1% chance for the former at this point it time. I'll do whatever I can to get that up to 0.2%

[unused0]

Multics achieved security by building it into the h/w. That caused the h/w to be more expensive and slower. The system market these days is all about price/performance ratio and the collective decision not to include security as a performance metric.

[mikewarot]

These days all the security is already in the hardware, that's what MMUs are supposed to do, virtualize memory, plus protect the OS from everything else. There's not really any penalty to be had for security... plus, think of all the stuff like Virus scanners you can get rid of along the way.

I'd settle for a hardware enforced write protect on a microSD port that I could boot from... at least then I could get back to 1982 levels of security.

[abdnafees]

Very compelling argument.

[v3ss0n]

ChatGpt infused timeline is not considered prime for me.