What we can learn from the Atlassian layoff video
Posted by ninetofivedev@reddit | ExperiencedDevs | View on Reddit | 114 comments
No tokens generated during the composition of this post
For those who have not seen the video, a former atlassian engineer recently posted a video going over his time at atlassian. The title is clickbait, this is basically a system design video about building a load balancer controller at scale:
https://www.youtube.com/watch?v=55pTFVoclvE
Now, few things to address because I've seen the other comments section of people who clearly didn't watch the video.
Nothing he said in this video is likely to be proprietary. It's high level system design. He probably doesn't have to worried about getting sued, it all seems like fair use. But IANAL.
Here is what we can learn:
-
Talented engineers get laid off. I know most people should know this, a lot of time layoffs target experienced engineers as they have higher salaries. But this individual seems like he knows his craft and I don't doubt he won't have trouble finding work.
-
You need to be considering new opportunities more often than every 8 years. This is not your father's career anymore. Once you hit that 5 year mark, YMMV, but you should really be dipping your feet into that market.
-
This showcases what SWE really is. It's not banging out code on your keyboard. It's not always building the main product, but something ancillary to the main product. Something that facilitates productivity on the team.
There is more here, but I have to cut this short.
roodammy44@reddit
Usually with these big layoffs it’s entire business units that get taken down. It doesn’t matter if you’re as good as Woz, if you’re in the wrong place at the wrong time then that’s it.
Quite a lot of life is just luck. It’s something you have to learn to accept because eventually it will hit you (especially in this career).
BananasAndBrains@reddit
And sometimes you can jump off the sinking BU to another.
Unhappy-Ladder-4594@reddit
Sometimes
EkoChamberKryptonite@reddit
Except that doesn't matter. I read about folks who were told their BU was mission-critical and core to the business by a VP or so in December 2025 and by February 2026, that same VP said- "Sorry, your BU was included in the layoffs". Orgs will gutt what they want to gutt regardless of your metrics.
Outside-Storage-1523@reddit
I kinda feel 90% in life is explainable by luck. Maybe even more.
Nick__of__Time@reddit
Very true - careers have real path dependency.
Which-World-6533@reddit
Oh no...! Jira might get worse...! Lol.
theschuss@reddit
But how?
nick125@reddit
They could start taking inspiration from Azure DevOps
dumbdog47@reddit
what do you not like about azure devops?
nick125@reddit
The "full screen" modals are my most recent annoyance with it, like when you open a ticket from a query. It doesn't update the URL when you open the ticket, so you have to copy the ticket URL from the link in the top left. The back/forward buttons in the browser will navigate the page under the modal rather than closing the modal.
nsxwolf@reddit
All these companies fail to understand how much of our jobs consist of cutting and pasting the ticket number somewhere. They go out of their way to make that hard.
mr_brobot__@reddit
It drives me insane there isn’t a big fat “copy ticket id” button on my company’s JIRA.
ok_computer@reddit
Gitlab’s little commit/branch hash copy select button was nice and should be standard for any tracking ID.
FluffySmiles@reddit
But metrics!
Krushaaa@reddit
What do you like about azure devops?
sharpcoder29@reddit
ADO is pretty great my friend. ESP compared to alternatives.
ritchie70@reddit
I honestly find the Azure DevOps UI much easier to use than Jira.
Although it is somewhat one of those “would you rather be shot or stabbed“ moments.
Head-Bureaucrat@reddit
And both are highly affected by customization. I've used both with minimal customizations and didn't mind them (again, still not pleasant but not terrible.)
Also uses customized versions that were just... Horrible.
thephotoman@reddit
Spoken like someone who's never used the alternatives.
Seriously, there are things way worse than Jira in that space. I have had the misfortune of using them and cheering for Jira when it replaced them.
Digital-Chupacabra@reddit
They will find a way to surprise us!
beeskneecaps@reddit
Make sure to hard reload to see what changed
Oo__II__oO@reddit
[AI Marked Ticket as Duplicate]
FluffySmiles@reddit
Made me laugh. Thanks.
joexner@reddit
Even more bad AI?
Eric848448@reddit
Sell it to Microsoft?
theschuss@reddit
Eh, MS data structures are at least semi logical.
serg06@reddit
It's a piece of shit BUT it did get noticeably faster over the last few years. Anyone else notice that?
zoddrick@reddit
Ive literally built my own front end for it with an mcp server to manage epics and tickets for my team. It has drastically improved our ticket hygiene and I've even got Claude skills that work against the mcp for implementing tickets.
arxorr@reddit
Less new features = less chance for them to make it worse
ninetofivedev@reddit (OP)
Honestly... I no longer care what issue tracking my company uses. I'm hooking it up to claude instead of fighting what is almost certainly a terrible UI.
Sweaty-Willingness27@reddit
Atlassian does pay pretty well, or did, anyways, according to glassdoor, so I think I can't blame anyone for riding that money train as long as possible. I switch jobs every 2-3 years or so because best internal "raises" I can do is end up with less buying power than when I started the job.
From what I know of the video (didn't watch it, just read other's summaries), yea, it seems like something you'd tell a prospective employer while interviewing with them, albeit in a bit more detail than is probably necessary. He's not revealing API keys or handing over code.
Nobody is going to look at this and go "OMG, they use SQS! I can disrupt the Jira industry!"
eightslipsandagully@reddit
He's based in Australia and atlassian were famously the highest paying tech company here (outside of HFC). Also famous for having a great culture, until the previous CTO joined 4 years ago from Meta
swamy7799@reddit
The observation about career mobility is thoughtful too. The industry changes fast, and periodically checking the market is honestly a practical form of career maintenance now.
Local_Recording_2654@reddit
8 years? 5? I start soft prepping \~ the 3.5 year mark so I can have offers in hand once my last 4 year RSU vests
casastorta@reddit
Oh, no RSU refreshers for you?
Deaths_Intern@reddit
Couldn't possibly be cause their attitude and motivation starts dropping just as they start scratching the surface of business needs
casastorta@reddit
Yeah, I’ve heard of RSU vesting cliff, but a few places I’ve worked at (I am not much of a job hopper) had yearly RSU refresh.
And yeah, you just become properly “senior+” around 3-4 year mark, when you become one of those people most other devs from most other teams start reaching out to you for help at different areas. It’s crazy that some companies cliff people to leave around that time.
HatesBeingThatGuy@reddit
The vesting cliff is because you are brought in at 50% salary for your level, and if you aren't 50% of the band or better by beginning of your third year, you will have a cliff where "oh no my pay went down", when the reality is your performance wasn't good enough to justify continuing at that pay.
Local_Recording_2654@reddit
Weird comment, you don’t know anything about me.
Local_Recording_2654@reddit
Pretty large RSU refreshers, but not as large as the initial grants. If I can hop and make +200-400k more I will.
gravteck@reddit
YMMV, but there are plenty of us toiling away at large non-tech companies for long periods of time. Obviously don't stick with a bad fit, but my gigs were 2 years, 11 years, and now 9 years. I had no problem getting top of the market comp with respect to my region and titles. However, my job in the middle was for a poorly run consultancy and lots of clients. I saw a million examples of what I didn't want to do, and that's one reason it took me so long to find an FTE position to commit to.
ritchie70@reddit
I’m at 24 years at my current gig. At this point unless one of our vendors wants me, I’m staying until retirement, which is within the next 10 years.
knightcrusader@reddit
Are you me? I'm at 18 and plan to stay here until I retire if they don't give me the boot first. I got 20 years probably though... hope I can make it.
purplecow8@reddit
Or until you get laid off. You don't know which comes first. That's the whole point.
ritchie70@reddit
Yep but my skills are dated and age discrimination is both very real and nearly impossible to prove.
Planning to retire in 5 years or so. Probably possible in 3 but riskier.
purplecow8@reddit
Valid. Good luck
TacoBOTT@reddit
Same here. 8 yrs at my current job and am getting great comp. Last job was also 8 yrs but I’m really enjoying myself and still learning a ton in my current role, which is also important to me. I do look for other jobs and interview pretty frequently but ending up turning down offers when I compare.
HatesBeingThatGuy@reddit
I'd like to dip my feet every 8 years, but my salary increases 25% a year staying put.
Relevant-Ordinary169@reddit
You’re anal?
Flimsy_Benefit_1207@reddit
Some of us just keep getting new jobs to avoid this crap entirely.
Existing_Station9336@reddit
Is that still a viable strategy? Especially with the way the market is currently?
Sweaty-Willingness27@reddit
I have about 30 years exp. Been having a tough time since about 2020, but I look for remote only. Also gotta factor in the ageism thing, but in the past, I never had an issue getting a contract/salary in the DFW area. Even been rehired multiple times at places I already worked at, so I guess my work is at least passable.
Now I can't really get an interview, though I've been mostly passive in my job search (an application every other day or so, here and there).
blipojones@reddit
Did you take paycuts to stick to remote? Im at 8 years exp and think im gona have to slash my salary to get back into a job.
Sweaty-Willingness27@reddit
No, but I was working remote 100% since 2014 and "hybrid" since 2012.
I suppose in a sense I took a paycut since I don't go for the leetcode interviews/jobs. I'm a Principal SWE at about $200k base, $240k with (full) bonus, which (in a low COL area) I'm perfectly fine with. Just wanting to keep up with inflation at this point.
Few-Impact3986@reddit
The biggest thing to realize is if you are remote and the company is not, promotion are super slow to unlikely.
vilkazz@reddit
Harder to keep/increase the tc but not impossible
Flimsy_Benefit_1207@reddit
I don't leave the jobs usually, just get new ones.
nsxwolf@reddit
I only see lower salaries now.
zdubbzzz@reddit
Yup, when I sniff a layoff I could potentially be a part of I dip. Rinse and repeat
rcls0053@reddit
I've somehow ended up this way, completely unintentionally. Every job I've had as a software engineer have been 3-4 years.
Shazvox@reddit
Ding, ding, ding! Keep switching jobs for higher salaries. Save and invest diligently. When you no longer can avoid a BS position, give em the finger, check out and live off investments...
pythosynthesis@reddit
Don't underestimate the cost factor. It's obviously true that with YOE comes knowledge that is impossible to get from a junior, BUT! In terms of sheer output, 2 juniors that cost less than the experienced dev will outproduce the senior. And that's what matters most of the time. Extremely developed and niche skills are often a hindrance in the job market. That's why you have very few experienced people and many juniors in a healthy environment.
Experience is expensive. And unfortunately very often not needed. So you can just fire the experienced devs, or many of them, and th company will keep chugging along just fine. I mean, look at things objectively. companies fire experienced devs and they keep going, nothing collapses, the CEO doesn't get fired and investors are happy.
It's brutal, but that's the reality. Adapt to it and thrive, or perish.
RabbitLogic@reddit
Doesn't matter if two juniors are out producing if they build the wrong thing. This is were experience wins, the senior engineer has hunches from previous experience about what to build to solve complex business requirements.
s3gfau1t@reddit
OP has such a bad take. I've been with the same company for over ten years and had to live with the shit decisions younger me made as the primary SWE. Sometimes it's lack of domain knowledge, sometimes it's premature optimization, sometimes a pure lack of experience. Things have worked well enough over that time...
As you say performance is measured in more than just throughput. I've had to work around past decisions I've made, and that takes time. It's really hard at a small shop to justify retiring technical debt, so things tend to compound over time.
This is the same as people who pant about how good vibe coding is. Whacking a demo together is not the same problem as building something that's a going concern, extensible, and has to handle real world traffic.
Mountain_Sandwich126@reddit
Don't agree with this. Companies are successful because of the work experienced engineers to get them to that stage.
Every company goes through short term incentive cost cutting / outsourcing so the c suite get their millions and then cycle back to quality due to the internal disaster left by numpties, requiring more experienced devs to unpick the "outputs" and bring some sense back into the systems.
Think watermelon reporting, all green on the outside, red on the inside.
If you think juniors with AI solve problems and scale out systems the same way experienced people can, then there is a gap of knowledge.
Note im not talking about shit devs who are 10x 1 year experience, im talking about the devs that made these companies what they are.
ninetofivedev@reddit (OP)
I don’t think this holds true.
Junior devs don’t know what they’re doing.
There is a reason a lot of places only hire senior devs. They don’t have time to wait for the junior devs to figure it out, they need someone who can build them something now.
Maybe st like your typical IT sweatshop, this is true.
snappy845@reddit
His liability rating went up astronomically. good luck landing another job after that stunt
ninetofivedev@reddit (OP)
What stunt?
StatusPhilosopher719@reddit
the clickbait framing probably cost it views from the exact people who wouldve gotten the most out of it, the load balancer stuff is genuinely useful but the layoff angle buries the actual signal
Abject_Parsley_4525@reddit
Not really on topic, but the layout of your post + your opening line and following it up with the magic 3 points + "I have to cut this short" makes me believe 1000% more that this post was created with the assistance of an LLM.
ninetofivedev@reddit (OP)
Yeah. I’ve made several posts over the last few months where people have incorrectly made that assumption.
I had to cut it short because I had more thoughts but my toddler was crying and it had already been sitting open in the background for the entire morning.
Going to say what I say to everyone else; your ability to detect ai is worse than you think.
Plastic_Monitor_5786@reddit
But what's the rush? You can just wait until you have time if there's more you wanted to say.
dnbxna@reddit
They're rushed to get a post out everyday...
happydemon@reddit
You had to cut it short. Maybe, now that presumably you have time, go back and add whatever you cut short? All 3 points are pretty vague and low information.
ninetofivedev@reddit (OP)
Feel free to leave your input. I’ve lost my train of thought.
happydemon@reddit
Right, okay. Have a nice day.
ninetofivedev@reddit (OP)
Oh so you don’t actually care? I didn’t think so.
happydemon@reddit
I read your post, I read all of your responses, and I gained nothing. I'm cutting my losses lol.
EntropyRX@reddit
It doesn’t matter whether is human or AI slope. It’s the low quality crap that bothers people. You ended a Reddit post with “I have to cut it short” as if it was a stupid corporate meeting. Don’t write if you have to “cut it short”. What to even bother, it just sounds like slope, regardless if you regurgitated that stuff or an LLm.
ninetofivedev@reddit (OP)
Bro criticizing me for not putting enough effort into my post and he called it slope twice…
Tasty_Goat5144@reddit
Thats what ai would say:)
obelix_dogmatix@reddit
so JIRA could go out of business? Don’t threaten me with a good time!
mikelson_6@reddit
This is just one guy and we don’t know the full story of his performance. He might be good with technical stuff but poor with people and struggled with pursuing business metrics. I don’t understand why Internet makes such a big deal from this video
soccerdude2014@reddit
I literally work at Atlassian and worked with him. He was great and always helpful and kind, even though I wasn't on his team/org and had to cross collaborate with his team.
The layoffs definitely skewed to impact those with longer tenure (they probably had above average base pay and invested RSUs).
aroras@reddit
The other aspect that stood out to me about the video was that he seemed to imply that he designed and built these systems from the first day of his employment on his own. The odds of that being true are very low. In all likelihood, the design decisions were made in collaboration with long time members of the team. If so, how much contribution did he have to these architecture? Was he taking cues from other more senior engineers? Is he describing the decision making of others? It's not a bad marketing ploy for him to release this video but, in my experience, complex systems like this are rarely the product of 1 or 2 individuals.
ninetofivedev@reddit (OP)
You’re right but also I’ve seen numerous people get laid off who were performing to a fairly acceptable level over the years.
My guess is he built something the company needed and they no longer had ideas for him to build things so he got cut.
mikelson_6@reddit
We can only guess, that’s why I think it’s not reasonable to make assumptions based on this video
ninetofivedev@reddit (OP)
I think he did a great job with the video. The constant cuts I find a bit annoying, but I know that’s a new trend to keep it filled with content and less dead air.
nunyabizzy101@reddit
Been at my current employer 10 years now. The higher my net worth gets the more I struggle to find the motivation to look for another opportunity and go through the whole palava of interviewing plus settling in and understanding how another company operates plus potentially risk finding out that it's actually toxic and needing to go through the whole process again. Would you at $3.3M?
arstarsta@reddit
I find settling in a fun experience. The problem is more it's frownd upon to switch too often.
nunyabizzy101@reddit
You must be the type to find interviewing fun too aswell.
arstarsta@reddit
Yes love the interview itself but scheduling time is a hassle when I'm already employed.
GoodMenAll@reddit
It means your salary is at the bottom of the spreadsheet.
nunyabizzy101@reddit
Not sure what you mean there?
Pleasant-Cellist-927@reddit
He means that by joining a different place, either taking a pay cut or joining somewhere whereyour current salary isn't the top end of the company, you will be last on the bean counter's list when time comes to conduct mass layoffs.
Not sure how much of that is actually true though, IME mass layoffs tend to target underperforming teams or departments, not individuals by salary.
TheBinkz@reddit
Same. Been at mine for 6 years and the whole song and dance in modern interviewing is horrible.
intertubeluber@reddit
Same. My current gig could be my last. Interviewing is so disconnected from actual work I have to ask myself - do I want to invest a bunch of time into interview prep? Even if I doubled my salary, market fluctuations have more of an impact on my net worth than any compensation.
ninetofivedev@reddit (OP)
Always exceptions. You know your situation better than anyone.
skidmark_zuckerberg@reddit
I think the one thing that sucks is that it’s genuinely better to stick around at a job for 3-4. For me at least, I’d like to just put in 20-30yrs at a company and call it a day. But obviously not realistic anymore.
arstarsta@reddit
Apple maybe.
hibikir_40k@reddit
You should be careful about what you wish for. Many of the lifers I know are just very stagnant poeple who have very little technical perspective. You know that when your staff engineer du jour has 13 years of experience just where you are, they might be "a little" narrow minded, because they only have practical experience in one environment. It's quite the hassle to end up having to convince those people that many of the decisions they've made over the years are at best weird, and sometimes just straight out mistakes.
skidmark_zuckerberg@reddit
I mean if you can spend your entire career at one company and retire from it, I suppose it doesn’t really matter how much technical perspective you have beyond what you need for that one job. Work is just work at the end of the day. If you can make an entire career at one place none of it matters beyond what you do for that job.
But that’s a pretty obtuse way to go about it considering the chance you can actually stay at one company for an entire career is practically zero these days. Personally I do 3-5 years, and then on to the next.
Idea-Aggressive@reddit
He sounds great. But there are also cases of great ppl who don’t do much after awhile. Not saying that’s him
Goducks91@reddit
I’ve it’s less great people who don’t do much after awhile it’s usually their soft skills that hold them back. They don’t see the business side of coding or they are hard to get along with.
Idea-Aggressive@reddit
I’ve watched the whole video and he seems to have good soft skills; not only that, but he talks about handling difficult social situations, and explaining how much he learned from dealing with that, which is a plus!
In 8 years of work, a lot of us lose the motivation to show up everyday and contribute the same way you were in the first 6 months.
I find this to be the case across many companies, the most senior and older employees had very poor performance.
It’s very easy to read just checking the GitHub contributions.
Windyvale@reddit
What do you qualify as productivity exactly? Senior engineers and older employees are not as easy to quantify as newer employees. They have a lot of knowledge and understanding that drives the more easily quantified productivity of newer and more junior engineers. They can also accomplish more with less actual “work”while avoiding many pitfalls that would destroy productivity of many people who would have missed these.
Of course there is a range and I’ve seen unproductive ones myself even after correcting for the above, but I’m wondering if you are accounting for that?
As for the guy in the video, after watching it I would say he has potentially a lot going for him. It was more like a resume in video form than anything. He marketed himself well and I would not have thought to do it this way.
Idea-Aggressive@reddit
It’s not about the title and experience. But sense of urgency. Some people are very competent and accomplish a lot, others just do the bare minimal.
Goducks91@reddit
Yeah sorry, I wasn’t saying this person in particular.
Majestic_Diet_3883@reddit
Ex aws folks lol. They seem to just drop after getting that faang bag
royboypoly@reddit
Great video! Really enjoyed hear him talk through his projects.
GoodMenAll@reddit
I mean after 3 -4 years, need to start to look, you get salary bump every year that company now need to get rid of you and hire someone cheaper.
HelloSummer99@reddit
I don't mean this is any negative way, but your third point showcases why some positions are more secure than others. Always strive to work on "core" features to the product that bring in revenue, not the auxiliary ones. They get laid off first.
ninetofivedev@reddit (OP)
The core ones are often less tech agnostic.
If I’m a platform engineer. I can find a job anywhere. If I’m a product engineer for a TMS system, I probably understand trucking and logistics really well, but I only know surface level how k8s and cloud infra works.