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

Inside Palantir

buzzfeed.com

Post reply on HN
Print view View original

281–290 of 319 posts

Re: Inside Palantir

#281

Post by Anderkent »

dschiptsov wrote: post #251
Anderkent wrote: post #235

Earlier quoted context omitted.

> BTW, the guys who went there for sweatshop positions are those who barely managed to graduate Bullshit. I work at the London office, most of the FDEs have firsts from cambridge/oxford/imperial/some other top european university.

Oh, come on. I have seen enough of consultants and witnessed a SAP takeover of a small retailer. This looks like as if a whole bunch of hotshots in $3k suits take you to a head office in, say, Germany, with a five star hotel, top tier restaurants and business class flights, while telling you that you are too successful and too rich to drive your own car, that only struggling, unsuccessful losers do it, etc. Eventuall…

> Btw, really smart top grads would never even come close to these scam shops

Nice attempt at moving the goal post; first you claim only 'those who barely managed to graduate' work at those places. Now it's 'really smart top grads' don't work there.

I have no idea how the rest of what you said is relevant to the article, but your claim is simply factually incorrect.

Re: Inside Palantir

#282

Post by ethanbond »

deprave wrote: post #231
_tpzm wrote: post #213

Earlier quoted context omitted.

Yes, it's consulting. Palantir has done an amazing job branding itself as a product company even tho it's mostly not, and for a long time, up to ~1000 employees in 2012, branding itself as a startup. A surprising number of my friends and classmates (Stanford '12/'13) ended up working there. Stanford emails out a jobs summary for each graduating class. For 2013 graduates the #1 most common job title was Forward Deploy…

On one hand I agree with you that borrowing a term like "sweatshop" to describe Palantir is ridiculous. The abuse of language not just in this instance, but overall, is ridiculous to both extremes. Having said that, I think the intention was to describe Palantir in relative terms compared to other companies in SV. With that in mind, then yes, Palantir is one of the worst places to work at among companies of the same…

I think it can be bad if you're not a specific type of person. I loved interning at Palantir and absolutely hated interning at Facebook.

There are people who legitimately enjoy super high intensity environments. There are people who legitimately loathe low intensity environments. If you're not one of those people, you're not going to have a good time at Palantir — this is not a secret. This is stated over and over by everyone. Shyam (head of BD) has a public blog post shaming the very notion of work-life balance. No one is tricking anyone, except candidates fooling themselves into believing everyone is lying and it's actually a chill place.

Ultimately, the sweatshop analogy is pathetic on this basis alone: you have a choice to work there or not. Working conditions aren't really the problem with abusive employment. There are people who take on far more life-threatening and limb-damaging jobs than a sweatshop.

The key distinction from those people and those who are abused is that they are free to leave and thus free to demand proper compensation.

No Silicon Valley company is a sweatshop. What a delusionally entitled way to see the world.

Re: Inside Palantir

#283

Post by pbreit »

callmeed wrote: post #60

My understanding is that Palantir is running an engineering mill/sweatshop using fresh out of school CS grads (particularly from Stanford). That much of their time is spent simply writing client-specific code that imports and cleans up data from disparate enterprise systems. That its not challenging, innovating, or exciting. Of course this is all anecdotal based on what I've heard from friends and a Palantir engineer…

It's sorta Accenture-style consulting but with a lot of pre-built proprietary tech to achieve a high degree of leverage.

Re: Inside Palantir

#284

Post by thedufer »

zpoley wrote: post #167
desdiv wrote: post #151

Earlier quoted context omitted.

>Palantir could easily achieve 97% of their results with tier 2 grads, but because they only take Stanford/Princeton/MIT/etc, they can justify their high prices to clients. Do we actually have any data on Palantir's hiring practices? Because my personal anecdata is that I have a friend working there and he's from a tier 2 school (University of Toronto, maybe even tier 3?).

Paysa ranks Palantir #20 in terms of tech talent [1], and also supports a large fraction of their workforce coming from tier 1 schools, as well as salaries [2] that can help attract top talent, even if the work turns out to be unsatisfying. [1] https://www.paysa.com/company-rank#!companies=Google:::Airbn... [2] https://www.paysa.com/salaries/palantir-technologies

That salary data is pretty misleading. The equity grants are almost entirely illiquid, and if you take those out the numbers start to look normal (for startups) or low (for big tech companies). Not what you'd expect to need to attract Stanford grads.

Re: Inside Palantir

#285

Post by curiouscat321 »

callmeed wrote: post #60

My understanding is that Palantir is running an engineering mill/sweatshop using fresh out of school CS grads (particularly from Stanford). That much of their time is spent simply writing client-specific code that imports and cleans up data from disparate enterprise systems. That its not challenging, innovating, or exciting. Of course this is all anecdotal based on what I've heard from friends and a Palantir engineer…

I feel like a similar argument can be said for any major engineering company. Most new grads heading to Facebook, Google and the like are doing grunt work. Unless you're working at a true startup, where you know everybody's name, you'll be doing some degree of grunt work.

Re: Inside Palantir

#286

Post by forgetsusername »

dbt00 wrote: post #206
CamperBob2 wrote: post #149

Earlier quoted context omitted.

Mass surveillance isn't about being able to spy on everybody, it's about being able to spy on anybody. That's basically what companies like Palantir facilitate. If you work there, you may be a bad person, and you should probably feel bad. I watched "All the President's Men" on Amazon last night. Through practically every scene, I couldn't help thinking, What if people like this had the technology to keep from getting…

I worked there for 7+ years, I'm not a bad person, and I don't feel bad. If you're opposed to a surveillance state, you should care deeply about preventing future terrorist attacks. I was someone who spent a lot of time thinking and caring about privacy and overreaching law enforcement in the '90s. When the original PATRIOT act was passed in the wake of the 9/11 terrorist attacks, they weren't based on a deeply nuanc…

>I'm not a bad person

When push comes to shove, your personal opinion on whether you're good or bad is irrelevant. Others might judge you differently. I've said it before; if we ever have a violent proletariat uprising, I think there will be a few people here who are surprised to see what side they're considered "on".

The HBGary leaks showed a slide deck of Palantir suggesting a few very underhanded tricks to kill WikiLeaks and discredit people trying to do "good" work, for money. How and why is that acceptable?

Re: Inside Palantir

#287

Post by vonnik »

amiraliakbari wrote: post #274
vonnik wrote: post #2

This article is much more interesting than its fairly predictable headline implies. While the secrecy of Palantir has served as clickbait for many years, there's actually real news in this. And the news is about the difficulties of scaling a services-heavy, on-prem software company that basically rents out forward-deployed engineers at 10s of millions of dollars per year. Especially when the software is an open-sourc…

> Especially when the software is an open-source stack Is there specific information about stack used by Palantir? I just saw a few videos presenting their products.

Apache Hadoop, Hive and Pig, among others.

https://www.palantir.com/wp-assets/wp-content/uploads/2013/1...

And actually, it's usually an enterprise distro such as Cloudera's CDH rather than Apache Hadoop itself, which is kind of a nightmare.

Re: Inside Palantir

#289

Post by jonathankoren »

Anderkent wrote: post #234
jonathankoren wrote: post #214

Earlier quoted context omitted.

While I'll defer to you about what people you know, I don't understand how that doesn't sound like any Palantir product you know. It's the core product. When it's sold commercially it's used to find people committing fraud, and those people could be arrested, put on trial, and if found by a jury of their peers to have committed the crime beyond a reasonable doubt, thrown in jail for a number of years. When it's sold…

> When it's sold commercially it's used to find people committing fraud, and those people could be arrested, put on trial, and if found by a jury of their peers to have committed the crime beyond a reasonable doubt, thrown in jail for a number of years. Um, and that's a bad thing how, assuming they really committed fraud? > When it's sold to militaries and intelligence agencies it's used to make kill lists. Militarie…

The grandparent said he doesn't know of any Palantir product that sends people to jail. That just doesn't make any sense. But to answer your question: no, sending people to jail who have been convicted of a crime after a fair trial is not a bad thing.

My problem with Palantir is that it's a surveillance tool that's used to kill people.

You can't pass it off as "just a tool", because this was its initial rationale, and what its initial customers do. All the commercial fraud products are just after the fact product repositioning.

The algorithmic kill lists I refer to go by the military term "signature strike"[1]. To create target lists, the government essentially data mines their survelince data use social network analysis (sound like any product you know?) and throws it through a machine learning classifier. If the classifier comes back "true", the person gets droned. The key thing to remember -- and this can't be stressed enough -- NO ONE IN THE KILL CHAIN KNOWS THE TARGET! It's just math.[0]

Now think about that when your test fail, or you find a data error, or the next time you review your classification model's performance.

If you work for this company, you owe it to yourself to understand what exactly this company is doing, and how its products are used.

Palantir isn't like any other Silicon Valley company. Google just wants you click on ads. Facebook just wants you to click on ads. Uber just wants you to drive taxis and get taxi rides. Palantir rationale is to help people kill people.

Maybe you were overwhelmed with the private car, the talk of vague cool stuff, the hint of cloak and dagger, and the 6-figure salary, but you owe it to yourself to get informed about the company -- and any company you're thinking about joining. This is triplely so when you're dealing with intelligence and military contractors.

If you're honestly a pacifist, I hate to break it to you, but you've picked pretty much the worst company in Silicon Valley to work for.

[1] Dont bother with Wikipwdia. It redirects to "targeted killing", which isn't precise. A signature strike is a specific type of assassination. [0] http://www.usnews.com/news/articles/2015/06/18/obama-cia-ret...

Re: Inside Palantir

#290

Post by jgalt212 »

_tpzm wrote: post #213
callmeed wrote: post #60

My understanding is that Palantir is running an engineering mill/sweatshop using fresh out of school CS grads (particularly from Stanford). That much of their time is spent simply writing client-specific code that imports and cleans up data from disparate enterprise systems. That its not challenging, innovating, or exciting. Of course this is all anecdotal based on what I've heard from friends and a Palantir engineer…

Yes, it's consulting. Palantir has done an amazing job branding itself as a product company even tho it's mostly not, and for a long time, up to ~1000 employees in 2012, branding itself as a startup. A surprising number of my friends and classmates (Stanford '12/'13) ended up working there. Stanford emails out a jobs summary for each graduating class. For 2013 graduates the #1 most common job title was Forward Deploy…

For me sweatshops status is largely about the sort of work/life balance afforded its employees. If your job pays you $400K year, but you work more weekends than not. I'd argue your job is a sweatshop.
Post reply on HN