Live data from Hacker News

Ask HN: Startup acquired by a large company and it sucks. What to do?

news.ycombinator.com

101–110 of 464 posts

Re: Ask HN: Startup acquired by a large company and it sucks. What to do?

#101

Tell me you never ran a large code base in production, without telling me. Past a certain point there are no "simple code changes" anymore. Automation and functional QA needs to ensure zero regressions, internal and external docs need to be considered, it all needs to be bundled in with many other code changes (feature and fixes), etc. There is no shortcut. Even FB had to dial back their "break shit in prod" mantra.…

> very grown up sport, more like running a country than building a house

We need to stop calling things "grown up" when what we mean is that they take a different set of skills. It's childish, and I think it's been causing harm in our industry.

Re: Ask HN: Startup acquired by a large company and it sucks. What to do?

#103
A lot of people have talked about the issues with this post, but I wanted to call out one specific thing:

    The first engineer I talk to doesn't even attempt to answer my question but redirects me to their manager. Ok, that's odd, I think, but whatever.
This is definitely not odd. The whole point of managers is that they're there to protect their engineers from getting off track with every random request that comes up. I understand that it may feel bureaucratic, but as an engineer, someone who insists on talking to me directly without going through my manager is always going to mark themselves as someone who doesn't respect my time and doesn't understand that they're not the center of my universe.

This feels like a rant of someone who isn't willing to put themselves in the shoes of the other team and understand the real complexity of deploying and maintaining large infrastructure systems over long periods of time. Any "small option" that the deploy team adds today is most likely going to be an option that they're still maintaining 5, 10 years down the line. As the saying goes—"an ounce of prevention is worth a pound of cure". In the best case, taking a few days to get the requirements right up front is (hopefully!) going to save months and months of cumulative man-hours down the road for maintaining the system later, or making future changes. In your case, if I understand right, your work got approved within *two days*—hardly an onerous wait time.

I'll admit that the "for compliance reasons, you can't see our codebase" thing is kind of odd, and it would strike me as a bit of a red flag, but I'll also admit that, personally, if I was in that manager's shoes, I wouldn't want you anywhere near my codebase either. Just purely based on the way that you treat other engineers' time as worthless & fail to consider the possibility that the business has broader priorities that extend outside of your own personal tasks.

Re: Ask HN: Startup acquired by a large company and it sucks. What to do?

#104
post #37

> So I reach out to the manager and ask what is going on. This is a simple task, I said. Why does it take an entire quarter for your team to deliver? He doesn't have an answer. Your simple task, which you think would only take a few days to implement, is probably one request in a long queue of requests that team is dealing with. That means they won't be able to start on it for a long time. That team probably set up t…

The amount of people on HN nowadays that are okay with getting nothing useful done is striking. Startup guy, you don't have to accept being part of a bloated hierarchy that rewards politics more than good decision-making. You could check out at the corpo job and start prototyping your own startup ideas. Or join another early stage startup where your performance matters. I'll be starting one soon - let's keep in touch…

if your startup ever makes it to any sort of scale, it will likely have the same problems -- they all do :)

Re: Ask HN: Startup acquired by a large company and it sucks. What to do?

#105
"Is this how it's like at all large companies?" Short answer is probably yes. Depending on where you are located, but with large corps you may end up with SOX among other things.

Over time all the tools will fine tune to all the new processes and the overhead will hopefully be reduced. If there is ambition for this.

Also, have in mind that many large security breaches happens when large corporations acquisition smaller companies and tries to incorporate them too fast.

Do not be too quick to switch job is my general advice.

Re: Ask HN: Startup acquired by a large company and it sucks. What to do?

#106
post #37

> So I reach out to the manager and ask what is going on. This is a simple task, I said. Why does it take an entire quarter for your team to deliver? He doesn't have an answer. Your simple task, which you think would only take a few days to implement, is probably one request in a long queue of requests that team is dealing with. That means they won't be able to start on it for a long time. That team probably set up t…

> Take this as a learning experience: don't assume that you are at a startup where people will drop everything to handle a request from you right away. This doesn't explain why they wouldn't accept the offer of work from /u/lopkeny12ko. I've been in the same situation multiple times. Sometimes I've been allowed to do the work and sometimes not. In the cases where I was allowed to do the work, I certainly took longer…

Away team work is standard at Amazon. Of course you still may have to go to office hours and sign up for busy teams.

Re: Ask HN: Startup acquired by a large company and it sucks. What to do?

#108
Hi!

I have engineer's background and very much still feel like an engineer, but for over 10 years I have been managing a team of up to 50 developers within an organization with several hundreds of staff. Let me try to explain the "whys" below. Once you see the wider picture you might look at all this with less negativity, but it's also fine if you find it unacceptable - there are valid reasons why people leave corporations to work on something in smaller scale and what you describe fits a lot of these reasons.

Larger companies focus much more on avoiding the risk of breaking something than on the speed of implementing new features. The painful process you are going through is probably a result of earlier experiences of having requirements dropped into the backlog by random people. In your relatively small startup team you did not experience it, but in large corporations just by pure chance you can get hundreds of requests from all over the place. Each of them has costs associated with it: implementation, testing, change management (in order to keep track of stuff for a chance of a team resigning en masse) and the virtual cost of the risk of breaking stuff in production. Especially in a project which seems to be widely used by other teams.

The 4 pages long form is there to make sure that you really need the stuff done - otherwise you would not waste your time to write all that.

The two managers approving the change are there to prevent stupid stuff from being worked on just because somebody sits next to the developers and has various "ideas" and "needs". They are not there to discuss your idea with you, but rather to make sure it does not go against the project's priorities, which are unknown to you.

The lack of R/W access to the repo is to keep control over what happens in the project. It's not that they don't trust you, it's that you have no idea what they need outside of your small feature. After few months they need to be able to know (more or less) why something is present in the code, regardless of you being there or not.

You are not allowed to speak to the engineers, because they will be working on your feature around the end of Q1 2020, so you'd be wasting their time now. They have a backlog of other stuff to do and won't remember your input after few months anyway.

What can you do? Notify your manager about there being a blocker for your task and find something else useful to be busy with in case your blocker will land below a hundred of other blockers from other people. Remember that your personal efficiency is not equivalent to organization's efficiency and learn to handle several tasks at the same time while waiting for blockers for most of them to be resolved. Do not mistake it with multitasking (actually working on multiple things simultaneously), it's just an efficient scheduling of available resource - yourself - in the context of an organization working in unison.

PS. For most of the my time managing developers I've been working really hard to make sure people from across the organization do not interrupt them in their work. Once you put yourself in the skin of the engineers you are not allowed to talk to you will understand that at least some of the stupid mechanisms you've encountered are there for exacly that reason. You should respect that as an engineer.

Re: Ask HN: Startup acquired by a large company and it sucks. What to do?

#109
post #82

Earlier quoted context omitted.

>Learn how the system works, and learn how to be effective within that system. And what if the system is corrupt? Do you advocate complacency? > but I also know for sure that new employees who try to bypass process and question the priorities of other teams, before they learn the ropes and build credibility and trust with other people, do not succeed. And what if the people who do not succeed with such bureaucracy ar…

Literally the point of a job at a for-profit company is to make money, yes, primarily for yourself and secondarily for the business. This is by no means moral or virtuous. But it is true , and because it is true, the people who admit the truth of it will be more successful in the sense that they will be working with the system, and the people who act as if it is somehow untrue and there is some greater virtue behind…

It's both funny and sad you've got to remind that to people who post on HN of all places...

Re: Ask HN: Startup acquired by a large company and it sucks. What to do?

#110

Earlier quoted context omitted.

The amount of people on HN nowadays that are okay with getting nothing useful done is striking. Startup guy, you don't have to accept being part of a bloated hierarchy that rewards politics more than good decision-making. You could check out at the corpo job and start prototyping your own startup ideas. Or join another early stage startup where your performance matters. I'll be starting one soon - let's keep in touch…

if your startup ever makes it to any sort of scale, it will likely have the same problems -- they all do :)

and you're lucky if you can actually tell what the real problems are, because mostly I find people are chasing symptoms these days.
Post reply on HN