Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

How to succeed at Meta

teamblind.com

Post reply on HN
Print view View original

171–180 of 211 posts

Re: How to succeed at Meta

#171

Post by ryandrake »

DesiLurker wrote: post #87
trasz wrote: post #45

Earlier quoted context omitted.

If you have no problem knowingly wasting 30% of your life making useless shit in exchange for money then… well, enjoy, I’m not judging, but please realize many people don’t feel this way.

btw its not 30% its close to 50% if you count waking hours. I have never understood this single minded drive for Moar money. I mean I get upto a point but after that it sort of becomes a game in itself. OTOH you have basically extinguished your own 'signal' for resources that may have some dubious utility in future.

To me it’s about gaining back time at the end, when I may be unable to work. Every additional $N/yr translates into one less year I need to work before I can retire. Even once all your needs are met, it still makes sense to further optimize TC because you save it and that translates into earlier retirement. I have a sign above my monitor that reads “The Goal Is To Not Have To Work”

Re: How to succeed at Meta

#172

Post by fhsm »

AlexandrB wrote: post #148
crazygringo wrote: post #66

Earlier quoted context omitted.

Yup, this is all big companies. And if the metrics are wrong, that's not your responsibility . The PM's and VP's that come up with them know that metrics are imperfect and have tradeoffs. Let them handle the tradeoffs. If you're impacting the metrics, then you're doing the job assigned to you.

I'm pretty convinced that there's no set of metrics that can effectively capture productivity and business goals without being gamed. It seems to me that the only effective management system is a "web of trust" style system where middle managers have a lot more autonomy to make decisions based on their own judgement. Unfortunately it's hard to legally CYA if something goes wrong with that kind of setup and it has oth…

Self destructing measurement:

https://en.wikipedia.org/wiki/Goodhart%27s_law

All else equal, easy to capture proxies make good metrics; however, the point is that all else will not be equal.

Also, the CYAed version of web of trust is a “360 review”. Going all in on 360s is a good way to unleashing reality tv dynamics in a previously banal workplace.

Re: How to succeed at Meta

#173

Post by sopooneo »

y-c-o-m-b wrote: post #29

This person is getting a lot of snarky responses, but all of this is relatively true at Amazon also and it was true when I was at Intel 10 years ago. This is how large companies operate once they surpass the innovation phase, so it should be no surprise to anyone that's actually worked in a FAANG and doesn't pound down the kool-aid. Interesting how there's comments on this person being concerned solely with "TC" as i…

If you have a moment, what is “TC” here? Technical Contributions?

Re: How to succeed at Meta

#174

Post by bskap »

sopooneo wrote: post #173
y-c-o-m-b wrote: post #29

This person is getting a lot of snarky responses, but all of this is relatively true at Amazon also and it was true when I was at Intel 10 years ago. This is how large companies operate once they surpass the innovation phase, so it should be no surprise to anyone that's actually worked in a FAANG and doesn't pound down the kool-aid. Interesting how there's comments on this person being concerned solely with "TC" as i…

If you have a moment, what is “TC” here? Technical Contributions?

Total compensation

Re: How to succeed at Meta

#176

Post by lokimedes »

sooyoo wrote: post #168
sjsdaiuasgdia wrote: post #112

Earlier quoted context omitted.

Span of control. As you get above that number of reports, the manager is splitting their attention to the point that hot/cold spots are inevitable. https://en.wikipedia.org/wiki/Span_of_control

So, many layers are bad, but large span is also. This means you are limiting both width and depth of the management tree, effectively limiting the total size of the workforce to ... a few hundred? That doesn't sound right.

Separation of concerns also applies to organizations. You may see huge companies, but where business areas, divisions, product lines etc. work in isolation just fine. At the strategic level, things are steered like a mix of governmental and portfolio management, effectively abstracting the differences between the lower layers away.

To study single-mission mega projects, the classics are the Manhattan Project, the Apollo missions and the first ICBM systems. The management of these projects were awe inspiring.

Re: How to succeed at Meta

#177

Post by Our_Benefactors »

uoaei wrote: post #146
Our_Benefactors wrote: post #106

Earlier quoted context omitted.

> But I would bet a lot of money that it's mostly people who spend frivolously and selfishly so that their kids go to good schools and have good opportunities, but that others' kids don't get access to the same kinds of on-ramps to success. Getting a higher TC does not take opportunities away from other people. What kind of communist thought is this? Line employees at these companies aren’t the ones appealed to in “T…

Getting a higher TC does not take opportunities away from other people. Spending those earnings on things that do not improve the commonwealth is what is being discussed in this thread. It really is remarkable how, every time this subject comes up, reactionaries can swing only at straw men. It demonstrates a complete misunderstanding of the arguments of their supposed opponents, which makes them seem naive at best. I…

Convenient excuse to claim your opponent doesn’t understand your argument. I do - you are advocating against people being able to spend as they see fit, by placing the “commonwealth” above the individual. It’s communism.

Re: How to succeed at Meta

#178

Post by hintymad »

Ancalagon wrote: post #158
mrandish wrote: post #59

Earlier quoted context omitted.

> This is how large companies operate once they surpass the innovation phase I heard someone smart (maybe Naval Ravikant) say on a podcast that this becomes prevalent in orgs which have two or more layers of management between real customer-facing work and business ownership (CEO, co-founder, CTO, etc). Having worked in several startups as well as F500 FAANG-ish valley tech companies, I agree. When a manager reports…

I don’t understand. Literally every company I’ve ever worked for, successful startup included, had at least an engineering manager reporting to a director reporting to a VP reporting to a senior VP reporting to a CTO. Is this not a normal or good way of doing things?

In AWS, many orgs have manger to senior manager to director to VP to SVP to CEO. It looks AWS is really well managed if startups have the same structure.

Re: How to succeed at Meta

#179

Post by woooooo »

seanmcdirmid wrote: post #104
woooooo wrote: post #63

Earlier quoted context omitted.

Google rewards engineering work in a vacuum, divorced from customer impact. This can lead to bad engineering incentives like valuing system complexity over the actual problem you're trying to solve.

How would they go about associating engineering effort with customer impact when they have a bunch of sales people, product managers, project managers, UX designers, and so on, in between them? Heck, any large company where there is a division of labor will begin running into the problem of relating individual performance to overall customer impact, and if you begin measuring everyone by that singular metric, you wil…

Customer impact, business impact, some other subjective or objective measure.

The point is that there's a thing you're trying to accomplish with a project and the engineering techniques should be a means to an end rather than an end in themselves.

Re: How to succeed at Meta

#180

Post by maerF0x0 »

bskap wrote: post #174
sopooneo wrote: post #173

Earlier quoted context omitted.

If you have a moment, what is “TC” here? Technical Contributions?

Total compensation

aka the sum of salary, RSUs/Options, bonus, and other cash like benefits (such as 401k match).

Always remember to state TC as the offer/granted value (amortized per year, often divide by 4) not the present value of equity, that's how it stays relatively comparable across time.

Post reply on HN