hola?

>>51416951
>Is this normal

yes

>or is this startup fucked?

yes

there some places that arent fucked in the head but mostly people do software very very badly.

start looking for one thats not like this.

how do you do that? at the interview ask them about how they do software. in detail. eg, what is there software process? how does the work get parcelled out? how is quality asserted etcetcetc

>>51417009
Hmm I think I'll do that when my agreement is up this fall. I really hate pushing out untested shitty code because I have no time to unit test anything :(

Hey OP, I'm an SE at one of the big tech giants, totally opposite position as you.

I think it's just the startup mentality because most of what you said is contrary to how things are run here. Documentation is crazy important, etc. Difference is I've been working on lots of internal tools so I haven't had the experience of focusing on the customer or getting something out the door ASAP.

Seems to me the whole Agile thing is well and good if you want to bust out a prototype, but the finished product, even the initial release, should be a hell of a lot more polished/documented than what you're describing IMO.

Just my two cents - don't be afraid to look elsewhere for work. If the environment is bad it ain't gonna get better unless you try to make it so. Cheers.

>>51417025

indeed. it makes no sense whatsoever to not test. they pay sooner or later.

and whilst i'm here, when you're looking, try and find some place thats doing an agile method (scrum, xp, kanban etc) those processes are explicitly designed to combat the crap you're seeing.

of course they have to be really doing them not just paying them lip service for it to work. but it does work. again ask at the interview.

>>51417050
>the whole Agile thing

thats the problem he's not doing "the whole Agile thing". what he's doing bares no resemblance to scrum or xp at all.

Any way you can standardize what you' re doing, or at least sub-components? If so, that seems like a way you could speed up both pre-and post deal phases. Might again be some additional initial cost, but probably very much worth it in the long run.

>>51417050
True. I think I will look elsewhere. In the mean time I'll try to push back on my boss about better documentation and testing.

Any way you can standardize what you' re doing, or at least sub-components? If so, that seems like a way you could speed up both pre-and post deal phases. Might again be some additional initial cost, but probably very much worth it in the long run.

What exactly do you do btw? Because I can't imagine there not already being software out there that does what you do.

>>51417071

I'm not necessarily adverse to that because I think lots of times people get too focused on methodologies and end up doing less actual work.

But it sounds like you're neither using those methods nor producing much work - or at least work of good quality.

You're a smart dude - you know how to make good, readable, well-documented code. You know what a stable product should look like going out the door and how to make it maintainable internally. If the company doesn't know how to do these things better than you... It's not worth being there. If leave any job where I felt like there was no opportunity to learn and become better, surround myself with people smarter/more skilled than me.

>zero documentation
is fucked
>little unit testing
is fucked
>founder is ideas guy
is fucked
>zero communication between team
is fucked

The fuck kind of startup are you working in?

>>51417096
I can't tell you what I do exactly for confidentiality reasons, and yes there are alternatives to our tool but they don't go nearly as deep and we do.

I do standardize, tested code and try to point other devs at it to use it but at the end of the day it's their decision to use it or not. Not my place to tell people how to code I feel like.

>pushes shit software to clients
>barely any testing
>no documentation

how the fuck are you still in business?

>>51416951
To be honest I am in the same position, I feel its because the management has no hitech experience.
Infact, I am a little worse, I have 0 unit testing. Just because speed and getting my boss happy and shit.
I want to quit but I have invested so much in to this start up and I feel my own inexperience and stupidity is the one of the main reasons.
I try my best though.
Bottom line - all start ups fail like this

>>51416951
Go along the flow, in a year of so you'll be one of those guys that may not be the boss, but that the boss can't affrod to kick out.

>>51417183
I already am. I've been here 3 years now and become a valuable member of the team. I'm the guy you bug when you need something done asap.

I just see the inefficiencies and pitfalls the company keeps making over and over and I worry about its future and wonder if it's worth sticking around for.

>>51417183
So whats the point of this? Even if you dont get kicked out you cant change anything. You are just one person

>>51416951
I'm basically in the same position as you. At a startup where I manage an android app which I develop solo. Now here for 2 years and always released new software without unit tests, testing basically comes down to me and my boss playing with the application before we release it. I'm not completly retarded and do the obvious tests (boundry values, states etc..). Yeah sometimes we get calls of shit not working properly, but our customers always preffered our adhoc implementation of features instead of waiting for months. And we always have been able to fix the issues within a day or 2.

Documentation comes down to writing comment lines in the more complex methods, so that whenever I come back to it I'm not oblivious to what it does.

Slowly I just made some stuff in my free time, to showcase for my boss so he also can see the benefit. Like I started out developing on the production server, and well guess you guys know the shit it can produce. I started putting up a copy of the server for development, after showcasing that to my boss he gave me permission to work on it in work hours. Same with version control and some other shit I can't think of now.

Although I am looking for a bigger company to work for, I absolutly love this for my first job. Since the whole process from design until release was almost completly in my hands. Also the initial app was developed in India, so it was fun to indentify shitcode and rewrite it.

So I would say to do the same, prototype something and let your boss see the benefit.

>>51417360
hahaha yeah Indian devs are fucking shit dude. We have an India team and they butcher any code they can't copy and paste from some where else. They're great for grunt work though.

I use to make new applications/ code in off time too make our lives easier. Every time I'd show it to my dev team I don't get any response. No one seems to give a fuck so I stopped doing it.

For example, I made a simple app to simulate our tool locally via window services, or an app to version our API in the click of a button or a function of asyc delete amazon S3 files as quickly as possible (we were having issues with that).

>>>/reddit/

>>51417435
Yeah some parts were terrible, but I only cleared up the worst parts. Rest i rewrite when needed since they just wrote everything with 0 possibility for expansion at all.

And that sucks, maybe you should try to get your boss a bit more involved, like a meeting once a month where you can shoot ideas and discuss the workprocess. After all, if you can convince him of something, the whole devteam has to follow.

If that doesn't work and you really feel like your stuck only doing gruntwork, then you might have to look for something else.