OP says they don’t think the engineers hired by doing leetcode hard problems (and screening out the folks who can’t) does not engender a good future for the org.
Your argument is “I found the interview easy, and hence it’s not that bad.” Just pointing out that perhaps OP thinks the culture you’re part of is not a good one. I’m not taking sides here but curious what this clarification means to you.
Your point about engineering not being part of the problem is another. It’s partly true - unless your technology is true main selling point (like openai or maybe google search), then sure engineering makes or breaks the future of the company. So in one way you yourself point out the commodity nature of what companies like Lyft do - given how many companies around the world do it as well now that is. Then the question becomes what exactly are you paying top dollar to these engineers for? Create 4 more unnecessary open source projects to make sure the engineers feel like they’re getting exposure?
Airbnb is the epitome of this. Like dudes, you rent out rooms, what are you trying to do creating open source ecosystems in data engineering for?
Elons original thesis of Twitters problems was absolutely correct, why you needed 8000 people to run that org (or whatever the number is for Lyft especially on the engineering side) is beyond me. Overengineered is absolutely the correct term in these cases I think.
My argument is not that 'I found interview easy and hence not so bad'. My argument is that in Lyfts case engineering hasnt been the problem with why its business is failing.
> Then the question becomes what exactly are you paying top dollar to these engineers for?
Again Engineering is one vector in a company's success but not the only one. It might well make sense for Lyft to pay well for good engineering talent there. Maybe not paying well might have doom'd them much faster.
Your take on OSS is also a naive one . OSS helps in off boarding long term maintenance costs (if the project becomes popular enough) and staves off bit rot. It helps attract talent , creates industry standards etc. Engineering is feature multiplier and for some companies it does make sense to OSS.
I’m not against OSS, just frivolous ones funded by companies who should have no business doing that stuff. It’s not news that orgs like Uber were indulging the development and release of such projects so they can keep the fairly jobless highly paid engineers happy. There’s no way you can make an argument that the development and maintenance of airflow had any real bottom line effect on their ability to make a home rental market place. The true technology companies don’t and never did release oss solutions that were not fundamental things they could do well and needed to do well. I’m again happy (maybe I’m not) that airflow exists thanks to airbnbs indulgence, but as a bottom line for the company it doesn’t make sense. Especially in the current environment where you now are suddenly forced to show real value for your spend on engjneeeing.
I am sad that Lyft isn’t going to make it through, not only cause of the engineers but for their contributions for OSS and blogposts. I got some of their work as inspiration for some stuff. However, I acknowledge that some part of this can be a product of a ZIRP period.
Your views were commonplace 3 years ago. Now, the idea of paying for free OSS development with expensive engineer time, at an unprofitable company, in the middle of a tech-recession, is indefensible.
I would be careful not to express luxurious views in future interviews. Saying that Lyft engineering wasn't swollen and engaged in silly side-projects flies in the face of the layoffs announced in the article we're talking about.
An interviewee expressing those views would be seen as out-of-touch, if not stupid.
> Elons original thesis of Twitters problems was absolutely correct, why you needed 8000 people to run that org (or whatever the number is for Lyft especially on the engineering side) is beyond me. Overengineered is absolutely the correct term in these cases I think.
Anybody who read the mythical man-months knows that adding engineers to a project gets be detremental if it leads to a point where the marginal added communication overhead gets higher without really needing more people to achieve the goal.
The usual solution in big companies for the overengineering problem is that people who don't want to fight the org self-select, drink lattes all day long, and let the few people who are needed to solve the engineering problems do the work.
>The usual solution in big companies for the overengineering problem is that people who don't want to fight the org self-select, drink lattes all day long, and let the few people who are needed to solve the engineering problems do the work
Your argument is “I found the interview easy, and hence it’s not that bad.” Just pointing out that perhaps OP thinks the culture you’re part of is not a good one. I’m not taking sides here but curious what this clarification means to you.
Your point about engineering not being part of the problem is another. It’s partly true - unless your technology is true main selling point (like openai or maybe google search), then sure engineering makes or breaks the future of the company. So in one way you yourself point out the commodity nature of what companies like Lyft do - given how many companies around the world do it as well now that is. Then the question becomes what exactly are you paying top dollar to these engineers for? Create 4 more unnecessary open source projects to make sure the engineers feel like they’re getting exposure?
Airbnb is the epitome of this. Like dudes, you rent out rooms, what are you trying to do creating open source ecosystems in data engineering for?
Elons original thesis of Twitters problems was absolutely correct, why you needed 8000 people to run that org (or whatever the number is for Lyft especially on the engineering side) is beyond me. Overengineered is absolutely the correct term in these cases I think.