Inside Palantir | Hacker News Reader

Inside Palantir

buzzfeed.com

569 pointsJimDash2145307 comments

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 met at bar.

Anyone with more insight know if this is accurate?

They also like to recruit from around Washington DC area schools (Univ. of Maryland for example). Saw them at career fairs while also recruiting there. They all wear cool hoodies and project this agile young startup image. It seems to work, there is a decent size line of young hopeful grads wanted to work there.

That's interesting. I forgot to also mention that, from what I've heard, if you are going to work for Palantir, their NYC office is where you want to be. That the team(s) there more direct interaction with clients and influence on solutions/architecture.

It doesn't hurt that Palantir's offices in McLean are reminiscent of the GooglePlex while the top defense contractors' offices look like those in Office Space.

Yap, I agree. They are doing everything right recruiting-wise. I think only Google and Facebook had lines longer than them amongst maybe 50 companies or so.

The office isn't in McLean anymore, it's been moved to Georgetown. Much hipper.

That sounds like a nightmare, traffic-wise.

More or less. Alden will likely have a very juicy part 2 in another month.

You hear quite a bit about Bay Area companies paying top dollar for fresh CS graduates from top ranked schools. I wonder what compels a Stanford CS grad to take a job doing non-exciting work with a company that prides itself on paying below market rate.

Maybe it's the code names?

My guess is the hype created through secrecy.

The stock options and perceived prestige.

There was a point where Palantir became the most prestigious place to work out of school.

I think this was around 2011-2012 timeframe.

In a tiny fraction of the schools. I hadn't even heard of them while working in a top research university at the time (this was Europe so maybe I was just missing out on a SV circle jerk).

In Silicon Valley, European schools aren't really considered among the top 10 engineering schools. I'm not saying that's right, just telling you what the perception is.

Possibly. We value US degrees pretty low in EU as well.

Uhhh what? ETH Zürich puts out some of the best CS grads in the world.

There's a reason Google has an absolutely massive office in Zurich. Anecdotally, some of the sharpest Googlers I have ever collaborated with are in Zurich.

True, but the number of non-us schools that any American in computer science can name tops out at like 20, and that's only if they've been grad school or recruiting long enough. Let's see how I can do at naming perceived good foriegn schools. I'm not going to google anything, so the any misnaming is an accurate representation of the data in my head. These are listed in no particular order.

North America and the Caribbean excluding the United States

- Canada: Simon Frasier, Waterloo, UBC, Toronto, McGill

- Mexico et al: nada

South America

- mas nada

Africa

- South Africa: I one time had a CS professor from the University of Natal, but that's all I really know about it.

Australia and Oceania

- New Zealand: Weka came out of University of Waikato or something. That's popular, so I guess it must have something going on.

Asia:

- Japan: University of Tokyo

- Taiwan: National Taiwanese University? That's shown up a few times I believe. Taiwan also has a Tsinghua, but that's not the good one, so don't count it. I on;y mentioned it because...

- China: Tsinghua University is very good. It's "the MIT of China." Peiking University is also a good school, but they're not as engineering focused. There's something called Beiha or something in Shanghai or so, that one's good. There's another school with a long name that's in Guangzhou or something that's on the coast but not near Shenzhen that whose colors are white and green that's good. Maybe it's Fujian?

- India: I know about IIT, but in all honesty I don't remember seeing any papers or anything like that come out of it in my field, so I'm biased against it, even if it's "the MIT of India."

- Israel: Haifa, and Tel Aviv

Europe:

- UK: Cambridge and Oxford are famous, but I've never heard anything about their engineer schools, so I'm going to put them in the impressing-looking-degree-but-actually-only-average-or-less pile with Harvard and Yale CS grads. (Sorry Harvard boys and Yalies. Should have gone to Cornell.), Aberdeen shows up from time to time, so that's a plus. There's Open University, which might be okay, but I don't really know. I know people that went to Southampton in HCI, so I'll say yes for HCI. Swansea exists, but I don't know.

- France: University 6 or something. Although, I think it recently changed it's name, so if you're using the new name, you're out luck, because I don't know any other schools in country.

- Switzerland: There's the school in Lausanne. The one with the french name, and the four letter acronym ECPI or something. The first word is Ecole or something. That's got a good CS and math program.

- Austria: Zurich

- Germany: There's the Max Planck Institute, but I don't think that's a school. There's a university in Cologne, but I can't tell you anything else about it. Other than that, I can't name anything, which is a real shame, because Germany is famous for engineering. I heard the all the schools took a nosedive after the Nazis killed the Jews, but I don't know if it's true or not.

- Finland: University of Helsinki was good enough for Linus, it's good enough for me.

- Russia: Moscow Technical University or something. That's the good math school in Moscow. but there's more than one university in Moscow, and so it's hard to remember which one is which. It's even harder, because I can't actually name the other schools. I think St Petersburg also has STEM school, but maybe I'm just making that up.

And that's pretty much the international STEM post-secondary educational system. So 25 schools that I'd call good. (Mentioned, but not good: Oxford, Cambridge, Cologne, Taiwan's Tsinghua, IIT (sorry India)) And even though Max Planck isn't a school, although if you have it on your resume, I'll be impressed.

I'm sure I missed a ton, but that's kind of my point.

I will say that, at least in CS academia, everyone has heard of IIT / Tsinghua, because it feels like 30+% of the PhD students came from there, so they are certainly held in high regard by CS people.

ETH Zurich is in Switzerland, not Austria, and "the one in Lausanne with the french name" is Ecole polytechnique fédérale de Lausanne (EPFL), the other Swiss federal institute of technology.

People in SV probably have heard of it, but don't know it well enough / haven't interacted with its graduates much. Same for Cambridge/Oxford (except that everyone has heard of them).

I think it's more that they're considered impossible to judge, and so they kind of get passed around with a blank "Hey, have you ever heard of this place?" Ironically, I think you do better if you're from China. There's enough Chinese immigrants to educate you about Chinese engineering schools. Europe is just too diverse to build up this

Either way, you end up not looking at what foreign school they went to, but rather what graduate program they got into in the US.

I interviewed there a few years ago.

I can tell you that at least at the time the impression given during the interviews was that of a deep, intellectually enriching place. Mostly everyone I talked to went to Stanford, Berkeley, Princeton, or a similarly ranked school. Many of those not straight out of school had come from Google or Facebook. The interview questions were among the hardest and most technically demanding I was asked. You're told of working on problems of massive scale involving true "big data".

On top of this, the budget for seeking out top engineering talent seemed to be without bound. I was flown out and put in a really overly nice hotel room. I was treated to a very expensive steak meal for lunch. I was just one of a very large group of candidates being luxuriously herded through their HQ that day and seemingly every day without end.

I had the exact same experience

My experience was similar, but with Facebook it was even more so. I'm not sure this correlates to the work you end up doing.

I had a similar experience as well. They put me in a very fancy hotel and let me charge anything to the room (don't worry, I absolutely did take advantage of this). On top of that, they sent an S550 to pick me up on the day of the interview...

So gimmicky and shallow, can't believe this works.

Yeah but I was a Sophomore or Junior in college and I remember feeling super awesome and important.

These kinds of things play with your ego especially if you're not mature enough to have ever experienced something like it before.

not directed at you, but more a frustration with human nature.

It's sales pure and simple, unless they're going to hand over the keys to the S550 after the interview regardless of outcome.

McKinsey flew me and a few other fresh grads to Dubai for a week because I was helping run the entrepreneurs society. 5* hotel, mock case study for 1/3 of the time and the rest tourism.

It was a holiday for the McKinsey guys running that thing as well, they appreciated being paid to go waterskiing and BBQing on the beach. For me it was luxury unlike anything I experienced until then - the rooms were so big we had to almost shout to be heard across them, and there was a sofa on the balcony! I took home loads of Hermes toiletries and cologne (I didn't have enough money on my bank account for the room deposit, and one of the consultants had to deposit with his card). In fact we even missed our Paris connection to Dubai, so McKinsey sent a consultant over to baby sit us, which meant taking a Mercedes to the Brasserie La Coupole for an epic feast followed by a night in a neighbouring 5* (I think in the Tour Montparnasse).

Unfortunately, I did not graduate with high enough grades (2.2 on the BA, cutoff was 2.1) for McKinsey, which is why they rejected me the first time. So by the third time they tried to headhunt me, I just told them upfront that their system/HR would reject me because of my grades and it was not worth our time continuing the discussion. They insisted, and it happened as predicted, making me wonder why they still care about that for more senior hires.

There was also a period of time in my life where I accepted conference speaking slots if the hotel was nice enough and close enough to my home. Can't resist a proper 5* hotel lunch buffet, especially if they have giant prawns and olive-roast beef and a decent beer on tap. I made my talks increasingly ridiculous, I think by the end I was explaining Starcraft cheese to middle-aged CTOs and quoting Delta Force founder Col. Charlie Beckwith before following up with droid-driven tractors (all relevant to ML in e-commerce, I promise). They loved it, got voted "best talk of the conference" a couple of times, not hard when 90% of the other talks were "so here's the 3Vs of Big Data" verbatim copied from Wikipedia.

> So by the third time they tried to headhunt me, I just told them upfront that their system/HR would reject me because of my grades and it was not worth our time continuing the discussion.

One wonders how a company as rich as McKinsey could have such piss-poor data management. Why not, I dunno, check that the person you're headhunting is already in your database?

Data privacy issues can keep head hunters from accessing such databases. Secondly, many, many recruiters (and consultants, too) are given company e-mail addresses to give the appearance of being an employee but are actually contractors that might as well be external recruiters.

Big, cash-loaded companies tend to enact policies more about protecting their wealth and reputation than to actually grow a company or even to keep it efficient. "Good enough" growth and revenue projections are enough to keep people happy and the bonuses coming in. Nobody goes into Big Four style consulting trying to actually change the world unlike at least a fair share of entrepreneurs globally.

You negotiated to keep the S550 and unlimited expense account, right?

Reminds me of "The Firm".

I agree they they spent a ton of money on recruiting, but many of my interview questions were absolutely terrible trivia questions. They were strictly questions that had a single answers regarding knowledge of shell command line functions. Not familiarity with usage, like when to use grep or ps aux, but things like how do you unzip and search with 5 keystrokes (zgrep)? For a new grad, it was a complete waste of time.

Secrecy, inexperience of the developers, and wooing them with money and a great interview experience. A solid first impression.

I quit Palantir after 6 months in 2011; I am a CMU CS grad and I worked at Google for 5 years before Palantir.

Their pitches of "We only hire the best and brightest" and "Our work truly changes the world" sold me on them.

Why did you quit? I'm curious what your observations were as an insider.

Bunch of reasons:

* Making hedge funds richer didn't actually feel like changing the world.

* I never really clicked on the codebase, didn't understand what was going on in the code, and didn't contribute much. (I don't think the code was easy to understand, and my mentor wasn't much help)

* I wasn't interested in making Palantir my full life, and there was a lot of pressure to have work be life. One of the leaders in my group basically said "yeah, I don't spend time with my pre-Palantir friends anymore."

* I also didn't click with my coworkers. I was about 27, and felt like an old guy. There were a bunch of other things socially that didn't work out. Strong culture that I didn't jive with.

* I got the signal that my 9-7 schedule wasn't enough.

The Government group may have been a better fit for me, I'm not sure.

Hey, not everybody there was baby faced. Your lead when you quit was 35. :)

Pretty lame pitch. Yeah help coca-cola sell more sugar water. I guess that does change the world.

Also best/brightest worship needs to stop. It correlates to nothing. This coming from MIT/harvard grad.

Right. As soon as I hear mission statement bullshit from a company, I plot the mental exit strategy. Amazing how those Jedi mind tricks work on the supposedly "best and brightest". I could see it working for a cancer research or fusion energy team, but work focusing clicking on ads or spying on people as a mission is bullshit that you cant polish from my perspective.

I got a pitch a couple weeks ago from unity inviting me to 'change the world innovating the best game development tool/ecosystem'

It's so tone deaf it blows my mind.

It helps that their primary competitors are explicitly about selling ads.

Palantir can at least pretend that your work will be "changing the world" since they do have some government and NGO projects with real impacts. Of course, only a very small select few will work on those projects—not that they mention that in recruiting.

The offer letter includes the stock/salary options based off of the current and projected valuation of the company. It's hard not to be impressed when you see the eye-popping values of stock options after they vest (4 or 5 years).

At least that was my experience in 2011. Had I taken the offer I would have had no problem affording a Bay Area house today, given Palantir's current valuation. That also assumes a liquidity event.

Do you regret not taking that job? (Granted it's hard to answer without having all the information available, such as how happy you would've ended up being there over the past five years.)

Yes and no, I chose not to take the job since my spouse didn't want to relocate to Virginia from central Maryland. The job I had at the time was fantastic as well. I have no doubt the position I would have fulfilled at Palantir would be intellectually challenging and rewarding. The salary was also competitive for the DC area (not at all for Palo Alto).

So on one hand I made my wife happy but who's to say buying a house in cash might have made her happy too.

My colleagues that went there for the most part enjoyed the work. One of them got burned out but I'm confident that was self-imposed and not due to excessive pressure from team leaders.

Oh, similar story! I used to live in Montgomery County, and I interviewed with Palantir precisely because it was a top-tier tech company that I wouldn't have to move for (albeit the commute wouldn't have been amazing). I didn't take interview preparation seriously though and flubbed the interview, so nothing came of it. I did end up relocating to NYC to work a different top-tier tech company that has an engineering office here but not really one in DC.

I've basically taken every good job offer I've had in my life, and I haven't regretted one jump.

That would have been a very big assumption http://www.cnbc.com/2014/03/19/free-advice-dont-go-public-sa...

I think Palantir has been overtly committed to avoiding an IPO for a while.

Yes, the "liquidity event" mentioned in the article is extremely vague. I'll reach out to my colleagues that are still at Palantir to find out what that actually meant.

Liquidity events are simply opportunities to convert some of the stock options to cash, at current FMV.

It's the ability to do board-sanctioned sales of stock to buyers. It's basically an equity follow-on sale to previous investors (or new investors if there's a round concurrently underway).

This is significant because private companies typically have clauses in their option agreements that you can only sell shares pre-IPO if the board approves. So, a company like Palantir structures things so that, before an IPO, you can only make money off your options if you're still working at the company. Oh yeah, and if you leave and want to exercise your options so they don't expire (so that you can have a nest egg that vests when they IPO), be prepared to pay AMT on the difference between the strike price and the current value.

They call it "golden handcuffs" for a reason.

Are you sure that Palantir options expire after the 90 days? The options offered to me were NSOs; I wouldn't have had to convert them.

Or are you making a general statement about companies similar to Palantir?

I don't think they've been that overt.

Here's one of the founders a couple years ago https://www.quora.com/When-will-Palantir-go-public?share=1

They've been signalling that an IPO is imminent for a long time.

Palantir has allowed for current employees to cash out options during some rounds of funding in the past, if I remember correctly.

I forgot to add that the actual valuation in 2015 exceeded the median projected valuation listed on that offer letter.

Palantir has great marketing.

Several friends turned down offers from Facebook/Google/etc. for (much) more money to join Palantir because they "didn't want to sell ads" and somehow thought that Palantir's work would be more fulfilling. The government and NGO work certainly sounds compelling (you're going to be finding terrorists and improving public health) until you realize it's a soul-crushing job mostly pushing low-level data around.

Ironic, considering Palantir's involved in the ad world, too. I believe their NYC office does quite a bit of advertising/marketing analysis.

A few years ago I sat in on a CS course at Stanford and was able to count 4(!) identical palantir t-shirts on the same day.

It was ridiculous, I probably have the picture somewhere.

Maybe because they recently had a recruiting event and handed out shirts?

I got a tshirt at the techto meetup here in Toronto a few months ago.

Palantir tshirts are very popular on Australian university campuses too in my experience, from recruiting events they run.

Palantir T-shirts are super convenient. I don't care at all that someone else is wearing the same t-shirt, if it means I don't have to go clothes shopping.

This would jive with a talk I heard at a conference from someone working on a collaboration at GSK with them; although that talk made it sound like there was some machine learning magic that was helping them structure their unstructured data; some of which is currently in the form of non-standardized handwritten logs at manufacturing plants that are just scanned into PDFs before being handed off to Palantir.

On a side note, the code name mentioned in the talk I saw for the project at GSK was definitely not Ribo (as stated in the article). I don't know if this is a case of GSK having a different internal code name than Palantir's internal code name, but GSK's code name was also based on Tolkien.

In addition to building dashboards, integrating data and maintaining a data pipeline made up most of what I did as a commercial Forward Deployed Engineering (FDE) intern. From what I've seen, this is what most FDEs do. Software engineers, business cultivation engineers, and people in other positions do other things, and I don't know what FDEs outside of the commercial umbrella do.

I think your assessment is correct, though most people I worked with really enjoyed doing that kind of work for many hours a week.

It's not for everyone though, and it wasn't exactly what I was expecting when I accepted the offer. For example, I was expecting that FDEs would be doing more analysis on the data.

What the hell is a "forward deployed engineer" and "business cultivation engineer"? You mean support and sales?

My guess: forward deployed engineer = implement the system at a client, business cultivation engineer = sales

A forward-deployed engineer is a fancy title for an engineer who is stationed permanently onsite at the client, usually with their own dedicated desk/office.

A business cultivation engineer is basically a sales engineer responsible for interfacing with and being a technical resource for the client's technical buying team (e.g. IT department).

> who is stationed permanently onsite at the client

Not really necessary. The distinction between FDEs and product engineers is that the FDE work is focused on a single deployment, whereas a product engineer works on a single product and is once-removed from actual clients.

Where the FDE spends time and what products he works on depends on the client requirements.

I was a "Programmer Analyst" at a consulting firm for almost four years, which is the same thing as a "Forward Deployed Engineer". Basically what it means is a consultant who writes software for the client, frequently involving on-site travel and face-to-face interaction with clients to gather requirements and design systems.

The impression I got after talking to Palantir on a number of occasions and talking to different engineers is that a FDE is a cross between a contracted engineer needed for staff augmentation and a software integrator. These are both jobs that are typically bid really low (most system integrators in the DoD space pay pretty poor salaries compared to the vendors of the products that are being integrated) and are not all that different in function from a role like a classic SAP consultant.

This approach really doesn't scale very well but given that the customer base really, really does not like to just take off the shelf software and wants to pay boatloads of money to customize software to their specifications, the software product is just a means to reduce friction and to lower the cost of integration compared to, say, building your own huge data pipeline with open source products from the ground up like oh... DCGS-A and its rebranding as ICITE http://nypost.com/2014/10/27/army-spent-5b-on-failed-technol... https://fcw.com/articles/2015/03/03/icite-faces-resistances....

My understanding is that the userfacing tool is still the ancient Java tool Palantir's been hawking since what, 2006? and that the several HTML5 rewrites haven't worked well [1].

Am I hopelessly out of date?

1 - https://www.youtube.com/watch?v=cajoFN6yjB0

Yep, that tool, Palantir Gotham, formerly known as Palantir Government, is a giant tangle of Swing/Spring dependency injected AbstractAutowiringFactoryBeans, but it is pretty powerful when configured correctly.

Most BI tools are based around tabular data and charting (i.e. Tableau). But Gotham is the best tool I've seen for exploring graphical (i.e. nodes + edges) data.

The only other thing in the space worth using is IBM's Analyst's Notebook. That's why Palantir (including Shyam Sankar, quoted in the OP) used some dirty tricks [0] to undercut i2, the company (later acquired by IBM) that developed Analyst's Notebook.

[0] - http://documents.jdsupra.com/3d375dfd-be30-471c-b346-bf434f8...

i2's claims are astonishing. Can't help but think that, had the nationalty of the firms involved been the other way around, the Justice Dept would have prosecuted Palantir and Sankar (which would likely have finished them as a government contractor).

I wonder how much Palantir paid to settle the lawsuit.

FWIW, we're doing it in the web, at bigger scale (GPUs in the client + cloud), and targeting Tableau & Excel levels of visual querying & flexibility: graphistry.com . A bit different and a longer topic, we're also more focused on modern data analysis & web workflows. Still only in closed pilots, but things don't have to be so bad :)

Is Stanford CS grads working in sweatshops really a thing? Glad I didn't go to Stanford!

We really dodged a bullet.

Note that that work doesn't require a top CS degree. 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. It's classic consulting - McKinsey/Bain/BCG could achieve 97% of their results with tier 2 grads instead of Harvard/Wharton grads at lower salaries, but you gotta wow the clients to win the jobs.

What strikes me is when the clients know this to be true, but don't actually care. They have a budget and they need to spend it! Looking forward to the day when the market catches up to them.

The way it works is like this -- consultants like these have to have a good reputation. They are effectively the sacrificial goats to absorb failure. If the project fails everyone can say "but we had the best of the best doing this, if they failed, surely nobody could have done it better". In that model it makes sense to hire the most prestigious candidates.

Also remember clients also have their own bosses, board of directors, investors and supervisors to answer to.

It seems with Palantir it worked well with CIA. If you ever sold anything to Uncle Sam, you know he loves to pay lots of money and then pretty often it ends up not being used or forgotten. Guess they tried that with a few big companies but found out that Coca-Cola is not as careless with its budget as Uncle Sam, and \had a hard time convincing them to just throw away tens of millions on some analytics thing.

It's part of the cowardice game.

How do you both take credit when you succeed and avoid blame when you fail?

You take an action that's easy to justify by external validity factors. If you succeed, you made the call, you take the credit. If you fail, you justify it using external validity factors ("they're one of SV's most valued companies! they only hire Stanford graduates!") to indicate there's no way anyone else could have done better.

ah yes, a textbook NMP [1], CYA [2], OPM [3] scenario. [4]

[1] NMP: Not My Problem

[2] CYA: Cover Your Ass

[3] OPM: Other Peoples' Money

[4] Profit!

It's not a budgetary issue. Generally the consultants are brought in to bring about a conclusion that the managerial team already decided upon. The expense and prestige of the consulting firm now backs this conclusion with sunk cost fallacy, and the managerial team has political leverage to push forward with their plan.

And, if it fails, political cover.

This.

People bring in consultants for one of two possible outcomes:

1) Rubber stamp a preconceived conclusion or idea;

2) Test out an idea and take the hit if it fails.

Basically, they are there to ensure a win-win scenario for the commissioning member(s) of management.

And to that end, the elite degrees confer a sense of authority and prestige. Additionally, "sounding smart" is a skill that such students have cultivated over the years in addition to but independent of possibly also being smart.

Huh, that got a bit convoluted near the end.

Think bulshit artist but with brains

Can confirm. Fake it till you make it. But with a really smart person who can fake extremely well.

>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?).

Sure. Here are 818 data points: https://www.glassdoor.com/Interview/Palantir-Technologies-In...

Top 5 schools according to LinkedIn are Stanford, Cornell, Berkeley, MIT, and CMU. See https://www.linkedin.com/vsearch/p?f_CC=20708&trk=rr_connect...

I have an additional data point of someone who didn't finish any university.

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.

Does anyone else find it a bit ironic that the same guy who wants to pay people to drop out of college is also the guy who only wants to hire from the "Top schools."

I'm old enough now to know where you went to school doesn't correlate to who you are as a person and your level of intelligence (however you may wish to define the different types of intelligence). That and I've met way too many people from "Top Schools" who are just this side of morons, and too many people who have zero or little education who are quite intelligent/innovative/great "engineers"/etc. Still, it's all about that brand.

The difference is Thiel is paying people to drop out of college to start a business, not to just go get a job.

But yes, there is irony, maybe not as much as at first glance.

> Does anyone else find it a bit ironic that the same guy who wants to pay people to drop out of college is also the guy who only wants to hire from the "Top schools."

From what I understand, Thiel is not super-involved in the day-to-day running of Palantir.

It's ironic that the guy who says college isn't best for everyone finds it a useful probabilistic filter, and a useful way to sell brand to clients?

How can I determine which school belongs to which tier? I always thought of University of Toronto as a top school based on the fact that it employs some of the best computer scientists in the world (Geoff Hinton is practically a celebrity nowadays). Apparently they were a third-world diploma mill all along and I never knew.

> How can I determine which school belongs to which tier?

Just out of curiosity, what is it that you do that you need to know this?

It helps with reading comments on HN that discuss tiers of schools. Shared context and all, even if the tiers themselves are meaningless (see Big 3 CS companies etc).

What ar the big 3? This seems like a meaningless distinction. I always see Big 4 on Reddit. I don't think anyone has ever said Bi4 4. And even Big 4 makes no sense. It should be Big 5. Wait or Big 10? Isn't that a football conference?

Microsoft, Facebook, Google, Amazon have prestige, but more importantly, recruit and hire the most heavily out of college.

These days, I think the big 3 are Google, Apple, and Facebook.

Amazon, Microsoft, IBM, Oracle, etc... are all great places to work, but are definitely on the backup-list of the best college grads.

I don't think Apple recruits as many SWE out of college as my original 4.

If we're talking about where college seniors want to work, it'd look more like a list of unicorn startups (Airbnb, Uber, and yes, Palantir) - reasoning is you actually get to build something instead of end up a cog in Google. But necessarily these places hire fewer college grads.

Yeah really, I was going to make the same comment. UofT is outstanding.

1745 current employees on LinkedIn [1] and 76 on AngelList [2]. Those are some data points that could be analyzed from their APIs.

1: https://www.linkedin.com/vsearch/p?f_CC=20708&trk=rr_connect...

2: https://angel.co/palantir-technologies

This applies to almost every tech and non-tech company, so this doesn't necessarily imply anything. All companies want to hire "the best" (let us set aside the totally valid question of whether or not elite schools graduate the best). And if you've done such a killer job at selling the reputation of the company to the public that all the top graduates want to work with you, why wouldn't you hire them?

So Palantir offers expensive custom-coded ETL? Awesome.

I'm sure it's awesome (not) to be a recent-graduate from a top university doing tons of mydata.this = theirdata.that, etc

I'm in kind of an odd position, because I worked there for many years with no CS degree at all and I was a lot older than most (not all). They do hire people who are great at CS things, and that often translates into mostly people from higher tier CS programs, but the filter that is applied during hiring is not that. "If you can do the job, we don't care what's on your resume. If you can't do the job... we don't care what's on your resume."

This is essentially a consulting company, which are heavy on name recognition and having pedigreed talent. It's pretty antithetical to the spirit of silicon valley IMO and is something much more typical of NY.

Or DC or Boston.

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 Deployed Engineer lol

Calling Palantir a "sweatshop" when the starting salaries are six figures is ridiculous.

There are 8 billion people in the world, and a lot of them actually do work long hours for a few USD a day with next to no rights.

Say you took a job at Palantir because you bought into the hype about fixing government. Now you work with some smart people and the pay and perks are pretty strong, but you find yourself doing consulting type work and feel unfulfilled. Maybe youre helping some large fund manager migrate off of Oracle DB or something.

The solution is to find more important work. Something you really care about. Something that directly helps at least a few of the 7.9b people out there who are less lucky than you.

Omg thank you for writing this. This industry needs to stop being so insular

> Calling Palantir a "sweatshop" when the starting salaries are six figures is ridiculous.

Why does the salary matter?

The labor in those firms are treated like fungible slaves, worked for illegal hours and disposed off uncerimoniously after a few years, to be replaced by fresh bodies.

"Sweatshop" is a mild term.

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 type: well-funded startups with a rather convenient revenue stream.

According to the arguments in this thread, Palantir is not a startup, so it cannot be placed in that type.

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.

>Stanford emails out a jobs summary for each graduating class. For 2013 graduates the #1 most common job title was Forward Deployed Engineer lol

This actually gives some credence to the sweatshop narrative. A Stanford education is so much about getting placed in a titled job that a manifest goes out every year about where graduates were allocated?

100k/yr as a rented contractor to a multi-billion dollar hedge fund does have a sweatshop feel. In-house programmers or independent contractors would be making much, much more.

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.

I think the hyperbolic use of "sweatshop" these days generally means "they throw labor at problems" instead of selling an innovative product / solution, rather than referring to low wages.

This is 100% what they are doing.

I know someone whose partner works at Palantir London. I've heard similar about the London office. Second hand though. What I've heard is, there is very little in the way of basic engineering practices or organization. People hacking stuff together all over the place, very poor test practices, stuff is broken all the time. Hero coding to get things working. And cult vibe through the roof. They weren't happy there.

https://news.ycombinator.com/item?id=10980506

This was my guess. The government is disentangling itself from the company, because IAD guys are starting to realize that Palantir is just another consulting firm.

Moreover, they're a consulting firm that failed to deliver on their promises. I'm not sure if they'll be audited and put to sleep, or if they have enough government patrons.

FinTech has realized Palantir's tech is pretty pictures and the rest is just a datalake setup. To negotiate and buy it is hard, even when it was the new hot company. Then to actually run the software be prepared to pay high rates for their "forward deployed engineers", which the rest of us just call "software consultants."

Their branding is excellent, in that they managed to offer a more "West Coast mentality," as I've heard sneered by government clients.

Having worked and lived with devs from the Bay Area, I can say that the the government could use a lot of help from Silicon Valley. But Palantir is not the help they need.

Questions:

What language are they using for cleaning up the data for import?

Is there a market for a middleman (i.e., something as a service) that cleans up enterprise data, using customized solutions if necessary, for import into different databases?

If this type of company already exists who are some examples?

This is very unsexy work which I happen to enjoy. Its the phase that routinely comprises 80% or more of any so-called "Big Data" project.

> If this type of company already exists who are some examples?

Palantir.

Data wrangling -- Trifacta, Tamr, etc. productized this so you don't need a consulting army.

Who are the clients and how much do they pay for these services? Whats the typical size of the dataset? Speed benchmarks shared with the public?

There is a company that called CloudFactory that offers a distributed task platform for data science. Data wrangling manpower on-demand.

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

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.

Hi all -- I wrote the Palantir article. It's really awesome to see all this discussion about it. As I say in the post, please don't hesitate to contact me if you'd like to chat confidentially. I am always eager to hear any tips or new information. Find me on WhatsApp, Telegram, Signal, or encrypted email. Contact info here:

https://www.buzzfeed.com/williamalden/inside-palantir-silico...

Do you have your PGP key fingerprint on your business card? If not, you should.

I don't! Good call.

By the way, your PGP key has no signatures on it. It could trivially be swapped out on your webhost if the server was compromised and no one would know the difference. You should go to a PGP key-signing party. I would offer to sign your key, but you aren't in the NYC office like I would've assumed. Fortunately there's no shortage of people to sign your key in SF!

This seems relevant right now https://moxie.org/blog/gpg-and-me/

Signal has a nicer communication user interface for some communication compared to email + PGP but it has not really solved the verification problem that many secure communication methods have.

The grandparent is advising the writer to improve the verification aspects that PGP provides.

Have you tried re-establishing trust with a Signal user who wiped their device? It falls back to the same PGP problem of comparing a string of numbers over a different secure channel.

Moxie addresses some problems with the OpenPGP RFC and with the GnuPG implementation, but after re-reading that post I don't see how it relates to the verifiability issue the grandparent is bringing up.

Or he could post his key signature on multiple media, provide links to them, and not expose his social network ;)

There's not particularly any connection between your social network and your PGP key signatures. That's what key-signing parties are good for: You get your needed signatures from people who are strangers, who are validating your identity through other means, such as access to your email account, physically matching photos that are associated with your various online presences, and possession of government-issued photo ID.

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-source stack, and the engineers are increasingly junior as the companies grows. That said, it's still great at sales. Big numbers there...

We took the questionable baity bit out of the title.

It seems to be perfect (and unfair to clients) business model. Same was described by Zed Shaw about ThoughtWorks.

1) Get expensive contract 2) Drop in junior (riding the fame that you only hire top engineers) 3) Train them on the job and bill the customer 4) Move trained consultant to work on internal project/high value stuff, replace with fresh blood 5) Repeat

> 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.

I want to see what Palantir actually produces. Like does Coca-Cola get a monthly report that says "Hey, you guys should bring back Cherry Cola in Montana for an expected 4% increase in sales" or what? What does $1mil/month actually get you?

https://www.youtube.com/watch?v=9ooAUeTlzdU

here's a quick demo of one of their products

edit: this is 2012, likely way out of date

Oh my god, what the hell is that thing! For all that money, they couldn't have hired a product management/design team?

Note the timestamp on the video... their products look sweet now.

> Ethan Bond

> Palantir, Product Designer; Starting June 2016, New York City, NY

I realize you're not technically a Palantir employee quite yet, but there's definitely the opportunity for bias here.

Yes, true. I removed the claim re: the quality of the team, hopefully it's a bit less misleading.

By the way, if Palantir's product actually works, your manager already knows you're talking about them.

That wasn't the concern. I said the team is very good relative to my experience at other places. They know that I like 'em already, don't worry ;)

Do you have any recent pictures of the products?

You can find some public stuff here: https://dribbble.com/palantir

I don't have anything else to point you to unfortunately.

I have to say, this does look pretty clean. Well done!

Looks like the web-based client work has finally started to produce a client. Nice work!

live example: http://www.cartercenter.org/syria-conflict-map/

Here is a live version of a [Palantir product for the Carter Center](http://www.cartercenter.org/syria-conflict-map/). I'd say it looks quite nice for a digital map.

Wow this is incredibly fascinating, how were they able to talk publicly about this...?

I can't take anyone who pronounces Iran(ɪˈrɑːn) "Eye-Ran" as a serious person.

Even if they had auburn hair and tawny eyes?

>I want to see what Palantir actually produces.

SAR. Suspicious Activity Report. If you don't know what "fusion center" is - https://www.palantir.com/wp-assets/media/capabilities-perspe...

Huh. Kind of looks like ArcGIS.

That's what I thought. ArcGIS advises you to deploy more Cherry Coke to Montana.

Don't ignore the network visualization on the right. My understanding is that it's a slice through an ontology driven knowledgebase. Palantir provides consulting services on the ontology as well as the integration services to map customer data to that ontology. The tool provides a visualization of this knowledgebase.

It looks like exactly the same sort of data you would get from a customized ArcGIS system or one of the BI (with GIS) software packages (such as Alteryx and the like).

But why pay $250,000 for software and also hire a good consulting firm when you can pay $18 million a year for the same thing. (I have just realized that I have been grossly underpaid.)

And I'm sure all those engineers they hire make excellent geographic analysts.

Actually maybe they do - they just brought back Cherry and Vanilla coke in Canada. I for one will be contributing to an increase in sales.

For what's it's worth, this type of analytics and market research is what Nielsen (NLSN) has been doing for several decades.. Indeed they were the first real BigData company. They just don't charge monthly.. You buy access to a set of data (markets, products hierarchies) and often some level of custom reporting and scheduled updates on the data.

Palantir is much closer to Accenture than Nielsen.

For the most part, Palantir is helping you analyze data you already have, not providing you with new data.

FWIW, Gawker reporters have been making FOIA requests of various agencies about their contracts with Palantir...you can see the responses on MuckRock:

https://www.muckrock.com/search/?q=palantir&models=foia.foia...

Many of those requests came back empty, here's one that produced responsive documents: the NYC Department of Finance -- $150,000 for a 6-month pilot of their "Perpetual Server Core" technology to do fraud detection:

https://www.muckrock.com/foi/new-york-city-17/palantir-contr...

edit: specified that the requesters appear to be Gawker reporters doing their own investigation

"Those anxieties come amid a wave of staff departures. A chart from Palantir’s internal wiki said the departures through mid-April amounted to 5.8% of all staff, or an annualized rate of 20%. That compares to a departure rate of 13.6% in 2015, 12.2% in 2014, and 9.2% in 2013. Palantir paid annual bonuses in March...."

Many companies have waves of departures in the spring. Bonuses are paid (as at Palantir), holidays are over, kids are about to finish school. A 5.8% departure rate in the beginning of the year cannot be "annualized" any more than fruitcake sales from December. And if it does end up that 20% of Palantir quits in 2016, that'd be totally normal attrition for a large company. Where I used to work it was 30% among software engineers, and even this was not a problem.

I imagine that a number of ex-employees go on to work in the industry whose data they analyzed at Palantir. This may help to explain why employees are willing to work for less than elsewhere. It's because three years later they will work for much more elsewhere. Especially the hedge fund analysts.

Saw the aftermath of a Palantir project after the client parted ways with them. What a train wreck. There was a lot of hype around them but when one looked deep into what they actually did it was a lot of smoke and mirrors and not a lot of substance. Even basic stuff like data cleaning and integration was poorly executed. Senior leadership didn't see value in what Panintir did, the project was cut off and they were asked to leave.

Not surprised in the slightest to hear they are struggling.

The problem with Palantir is the following:

Three letter agencies are paying and will continue to pay for Palantir consulting and products.

However Palantir has been unable to fully productize their solution (it is still pretty much consulting). Thus they have hard time convincing Fortune 500 companies to pay due to the costs and depending too much on human interaction.

> unable to fully productize their solution

In my experience with this type of problem, it happens due to a failure of vision and leadership. The software company focuses a vast majority of resources on POCs and "consulting" revenue (really, rent-a-coder revenues) and these "consultants" end up failing as product innovators (i.e. they contribute practically nothing to the productization or even harnessing of institutional knowledge) and they also fail as true consultants (i.e. they do not have deep industry experience or management consulting prowess to be able to deliver unique value and engage with clients). Ultimately, this happens because leadership neglects product and becomes too caught up chasing deals and burdening the organization with custom development work to close the gap between the existing product versus market needs.

This is OK if your software company's consulting services are affordable. Enterprise clients tends to be quite accepting of the need for customization after product purchases. However they do not expect to pay McKinsey levels of fees (let alone $18M per year?!).

An alternative business model that seems to work is management consulting, where you can charge huge consulting fees (if you have earned the requisite reputation), and you can even cross-sell relatively low cost in-house software products or services alongside the consulting work. However, high value consulting engagement like McKinsey tend to last weeks (not years), making the expense palatable to clients (e.g. around 500k fee, for 2-3 consultants for a month or so). The difference between management consulting and enterprise software is that when paying $200k for 2-3 "millennial" "management consultants" for a week of work, the client doesn't expect delivery of a complete working software solution - the client only gets a PowerPoint presentation suggesting what and how management should do in order to achieve some given goals.

There is an interesting potential disruption occurring at the intersection of management consulting and software, but not much has changed yet with traditional business model divide between old-school management consulting and enterprise software, and Palantir doesn't seem to have cracked that code.

> The software company focuses a vast majority of resources on POCs and "consulting" revenue (really, rent-a-coder revenues) and these "consultants" end up failing as product innovators (i.e. they contribute practically nothing to the productization or even harnessing of institutional knowledge) and they also fail as true consultants (i.e. they do not have deep industry experience or management consulting prowess to be able to deliver unique value and engage with clients). Ultimately, this happens because leadership neglects product and becomes too caught up chasing deals and burdening the organization with custom development work to close the gap between the existing product versus market needs.

Wow, that's a good summary. It's a trap that companies fall into and can never get out of:

1. Company has Product X and a bunch of smart engineers excitedly working on it.

2. The sales process ensues. No customers actually want Product X, but a few whales say "We don't want that, but you have smart people, we will pay you for this sorta-related ABC custom solution for us!"

3. Company tries to stuff ABC into Product X. They think they are improving their product AND getting the customer to pay for all the NRE! Woohoo! But in reality they've just turned into a contract engineering company and don't know it.

4. Steps 2 and 3 repeat until Product X is a franken-product that has so many features in it, nobody knows what it is supposed to do, customers still don't want it, and the engineers eventually realized that they are no longer working on the product they were so excited about.

"Enterprise" Management Consulting (with code) is what kept ringing in my head reading this article.

You hit the nail on the head, though I would also highlight this from the article:

>But Kimberly-Clark was getting cold feet by early 2016. In January, a year after the initial pilot, Kimberly-Clark executive Anthony J. Palmer said he still wasn’t ready to sign a binding contract, meeting notes show. Palmer also “confirmed our suspicion” that a primary reason Kimberly-Clark had not moved forward was that “they wanted to see if they could do it cheaper themselves,” Kelt told colleagues in January.

My own experience (small, not a big-5 at all) with Management Consulting, and following the sales and execution process, was many times this issue. Providing enough information and value to get FURTHER engagements, with many clients seeing "20% increase in savings is better then $xM a year to $firm" (our little firm).

Additionally, our company was used as a cudgel against procurement contracts, such as telecom, wherein, if we are brought in, perhaps the 5 year $30M telecom deal would create a cost decrease of 45-55%, so, the incumbent immediately shaved 20% off of the renewal contract, and the company did not engage with our firm further.

telecom and IT procurement savings normally dealing with CFO's/CIO's at $1b-$5b companies. Engagements from 3-8 months standard, with a few at several years. Every contract was a big fight of the above push and pull.

I don't think this is all that surprising. If you look at their staffing numbers and their fundraises, they appear to not be sustaining business but just growing staff because that's what startups do.

> Lisa Gordon, said that “the majority of the company’s customer relationships are multiple years in length, and many are as long as 10 years.”

This is not good at all. Customer loyalty is important, but considering that they've been adding staff at a fast rate, most of their customers should be new-ish.

They've raised about $2.5b over an incredible number of rounds (multiple per year), and are currently bringing in about $420m.

Based on my understanding, they message as a software seller, but appear to make most of that revenue off of consulting and integration services tied around their software lock-in. They also message as a big-data company, but AFAIK don't provide anything that would be called "big data solutions" these days.

edit: I thought Sankar's name sounded familiar, turns out he was the guy at the very heart of Palantir's very embarrassing industrial espionage and racketeering efforts against a competitor. He was apparently punished by being promoted to company president.

http://venturebeat.com/2011/02/16/palantirs-third-black-eye-...

They hire like a consultancy. Your topline scales linearly with the staff you can ship out to client sites.

What's interesting is that despite that being the case, and they charge out their staff at huge rates (especially compared to what they pay them), they're still losing money.

I don't see this as particularly damning to Palantir. From what I understand, Palantir is really like a McKinsey/BCG/Bain that specializes in data rich projects. If you view Palantir as another consulting firm, I'd be curious as to how its rates and deliverables compare to the that of MBB. My uneducated guess is that what Palantir is offering for its billing rate is probably in line with the standard for consulting firms.

But it is pitched to investors and clients as a product company, not a services company. The two get valued very differently.

That picture of Alex Karp is amazing.

If I ever get too rich to never ever give a f*ck, I will dress like that.

https://img.buzzfeed.com/buzzfeed-static/static/2016-05/6/12...

The article hints at a larger problem that underlies all data-science-as-a-service outfits: pricing for profitability. How can you price the generation of insights so that you; Palantir etc. can become profitable, while those insights can save on costs for the respective clients? Coke would need to generate $18M+ per year from the insights alone to justify the costs.

that $18M per annum price tag is kind of steep. at that price, if a consortium of five companies pool their money together, they could have a joint venture of a "closed" consortium without sharing data with outsiders. the valuation the new joint venture would even be another unicorn, depending on the multiples you choose.

Confusing company on many levels. On one hand such huge clients, top talent (until recently), and Thiel's famous success pre-requisite of having unique offering. But then ... they go and do really, really weird stuff like making recruitment videos shot on someone's Iphone: https://www.youtube.com/watch?v=PhMqPoCQ5Q8

I think for a long time they got away with seeming really cloudy and potentially mis-managed because of the secrecy that is inherent to what they do. But clients are now asking tough questions about what value they can actually bring to the table for crazy dollars, and things may becoming home to roost.

Seems like hardly fits the "competition is for losers" claim. Consulting firms are all about overpriced bullshit analysis. Seems like Palantir plays that game but sadly not even well.

I have no problem with that recruitment video. When you have the reputation of a secretive, shadowy company, it makes sense to forego a polished video and instead aim for something more organic and "real".

People want to work at something that mirrors Google, not LexCorp.

Spinning the reported facts another way, Palantir has doubled revs over last year, and could turn a profit at will. Meanwhile it is able to set price optimally at verge of pain point for some of the largest enterprise customers. Sounds like a well-run company.

> “One of the things we did well early on was to recognize and invest in the unique talents of each Palantirian”

I find it a bit weird how tech companies make up names for their employees now. Googler, Palantirian... does this make people work harder because they feel like they belong or something?

it appeals to that primitive instinct of belonging to a pack, fostering an us-vs-them mentality. this way, you are less likely to leave and are encouraged to give your 110%

It's also not just a tech thing, in consulting you have your McKinseyites, BCGers and Bainies

A shortcut to avoid calling them "Palantir employees"?!

I think in this situation, it's the less nefarious explanation that probably reflects reality.

It's probably just a shortcut for "X-company employees"

Internal reports are always written in Clingon!

Is it an American thing? We name folks from counties and states with nicknames too - Hoosiers from Indiana for instance. And folks from Cedar Rapids are Cedar-Rapidians around here.

It's not just an American thing, it's the concept of a "demonym" -- and there are official and unofficial demonyms. For example, the official demonym of Massachusetts is "Bay Stater," but everyone uses the far more popular "masshole."

There seems to be a lot of confusion here so let me clear up some things from my former several years at Palantir.

1) Palantir's product is a legitimate product that works quite well for many graph analysis use cases. If you can model your problem as a graph, chances are that Palantir will help you find some solid insights (if there are insights to be found) across your 15 disparate datasets. Software Engineers build the core platform.

2) Forward Deployed Engineers customize and extend the platform for specific client use cases and get client data into Palantir. If these customizations end up being useful, they get rolled into the core platform(ideally).

3) Palantir works their engineers very hard and the performance bar is very high. Calling it an engineering mill/sweatshop is a bit much considering you get 3 meals/day, massages, chiropractic, nap rooms, excellent events, laundry, etc and 6 figure salaries or massive stock or both.

4) Lots of people have been leaving. Part of this is because lots of oldies are hitting their 10 year marks and part of it is because Palantir went through a massive growth spurt over the last 4 years or so and now the fat is being trimmed/people are leaving after concluding that it wasn't a good fit/they don't want to work that much. If you can get into Palantir, you can get into Google: Why not go to Google?

5) If you want to see the actual product, just look around. Here are like 150 examples: https://www.youtube.com/playlist?list=PLCA98B156F7EFD6A0

6) The bookings vs revenue thing is pretty deceptive but everyone knows that the money isn't real until its in your bank account or there's a signed contract for it to be in your account.

From your description, this is a classic Accenture-style (and Bain and other strategic) consulting and systems integration business model, though possibly with higher quality work, based on that latest management buzzphrase Big Data. Whether it can provide the perceived value of a BCG of McKinsey is another matter.

At least they have the decency to tell you who they work for right in their name.

It always kind of amazed me as a pick for a company name. Like, they realize palantirs are a tool of the bad guys, right? They're not a good thing! It's like naming your company "Stormtrooper", or "Dementor".

Actually the palantiri were created by the elves as communication tools, even before the sun and moon were created, and used by the good guys for several thousand years before they fell in the hands of the bad guys about halfway through the third age.

Just like how a lot of computing was built by the good guys as communication tools/for fun, before the sun and moon were created back in the 1970's, and used by the good guys for several decades before computing fell into the hands of the bad guys aka surveillance state.

I've always interpreted it to mean that Thiel dislikes bullshit, so he's kinda sheepishly telling everyone who is sufficiently detail oriented/nerdy exactly what his new company is all about, right there in the name.

The explanation I've seen is that a palantir itself is value-neutral. However, because Sauron and Saruman each had one and used them regularly, using any of them became risky, because they were a channel through which information and influence could leak.

But that, in turn, is arguably a great metaphor for the chilling effects of advanced persistent threats and abuse of surveillance capabilities by our ostensible/former allies.

There's a company in Boston called Vecna [0]. I always thought it was odd to name a company after a famous evil lich [1]. Though now that I'm reading the company's About page, I see both names were probably inspired by the Czech word for "eternal".

[0] http://www.vecna.com/about-vecna/

[1] https://en.wikipedia.org/wiki/Vecna

I think it is happening unconsciously. Like say Chertoff named his naked-airport-scanner company Rapiscan. I think on some level they do it because they can, and that serves as a signal -- we are so awesome we can even use a name like that and still be successful.

Palantir is probably worth less than the money invested. At $420 million in revenue, they would need to be worth more than 6x revenue to match the $2.5 billion invested. They certainly have "special sauce" over and above their value as a consulting firm (which are worth maybe 2x revenue).

But how big is the market for that "special sauce"?

The intelligence customers are low-capability, have massive datasets and really do need Palantir. But they've tapped out that market. So look at the consumer brands in their customer list: low capability, yes. But you could fit the data for any of them into the RAM on one server.

Yes Palantir has super smart guys who can find fascinating relationships in the data. But there are only so many relationhips to find. And once that's done, the IT staff of the customer can do the work easily themselves.

So I'm sorry but I think Palantir is a washout (worth less than the money invested). Nice offices though.

I'm actually surprised that Palatir paid below market rates. I heard they paid well above market rates, which justified their stance of never going public.

I hadn't heard that either. Their reputation in the DC area is that the generally pay market rate, although I think part of their recruitment strategy is that they're less stodgy than the other big defense contractors. (I.e. "we're a software company in the defense/intelligence space, not a defense/intelligence company that does software".) Not sure if that's really true as I've never worked there, but that's the pitch.

What I've heard is that they pay well for incoming grads, but then no pay raises...ever...and by year 5 you're clawing at the walls for a liquidity event so you can shed your options and get out.

Whatever you do, please don't discontinue Plottable.js

Wow, that is one great web site.

Plottable.js is the only charting package I've used that's a real library, not just a set of pre-canned charts. Sadly, the devs don't seem to have a lot of time to devote to updating the site.

It's open source.

$1 million/month. I think that would make even patio11 blush. :)

Raise your rates!

20% annualized turnover does not seem alarming for a company in the valley that hires mostly fresh college grads.

That implies an average length of tenure of 5 years. Very few college grads that I know stay at their first job for 5 years.

The average doesn't tell the true story tho'. If you have a few old-timers in management but a majority of engineers churning in a year, then your product is mainly being worked on by people who never get fully up to speed on your codebase, supervised by people who haven't programmed in years.

"The company, based in Palo Alto, California, is essentially a hybrid software and consulting firm, placing what it calls “forward deployed engineers” on-site at client offices."

This says everything about Palantir, and depending on your perspective, their valuation is (not) justified.

It's justified if you think of the equation Palantir = McKinsey + IBM. On the other hand, no consulting firm should have 20x revenue multiple.

I've often heard the Palantir narrative where they talk about a core platform or product that they build independent of the consulting work. That might explain the optimistic 20x revenue multiple. However, I don't think that platform has really come to fruition and they're still relying primarily on consulting to fuel themselves. Someone else here mentioned that they hadn't fully productized their solution, which prevents them from closing large clients outside of government contracts.

How can they ask for such high prices?

By having connections to people spending someone else's dollar with low accountability.

Price discrimination at its finest.

> "If you can pay, you have to pay us. If you are a philanthropy stopping pedophiles, we will give you the same product you would get if you are wealthy special forces unit in a very wealthy country," Karp said.

> "If you're a hedge fund and you make a billion dollars and say you're not going to pay us, well, that's rapacious and we're not going to take your calls anymore."

http://www.cnbc.com/2014/03/19/free-advice-dont-go-public-sa...

To get an idea of the kind of stuff Palantir do then it's worth taking a look at the presentation that was leaked during the HBGary leaks: http://www.businessinsider.com/palantir-wikileaks-2011-2?IR=...

I used Palantir heavily in my previous work and it was disappointing at how manual it was based on what we were sold as a mostly automated platform.

It had a great interface for creating network diagrams, and collaborating but it was a huge pain to get integrated.

I mean if nothing else, give us keyword matching and linking!

Interviewed with them once: took like 2 months and it was a horrible experience. Weirdest thing was I randomly ran into this guy onsite that was a friend-of-a-friend.. and enough of an acquaintance that we had each others' numbers and went drinking a couple times... saw him there wearing the same Palantir t-shirt as everyone else (which was weird) and he barely acknowledged me... like I wasn't "one of them" yet... it was creepy enough that I was glad I didn't get an offer.

FYI:

What does Joe Lonsdale think of BuzzFeed's Inside Palantir article?

https://www.quora.com/What-does-Joe-Lonsdale-think-of-BuzzFe...

The headline graphic really caught my attention, because my fingers have ls -latr ingrained into them.

I thought it oddly fitting.

I know a few people who have worked for Palantir. They all agree it doesn't work. They all quit because Palantir data analysis is being used to ruin lives (throw people in jail) and they believe the whole operation to be a giant defrauding of the government.

I worked at Palantir for over 7 years, and that doesn't sound like any of the projects I knew about or any of the people that I worked with.

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 to militaries and intelligence agencies it's used to make kill lists.[0] Don't delude yourself into thinking that it's founding wasn't to algorithmically create kill lists, because it founded in 2004 with funding from the CIA specifically to bring "big data analysis" to the "War on Terror".

It's a scummy company. And if there's any doubt don't forget, they were implicated with HBGary in a dirty tricks plot against Wikileaks and Glenn Greenwald after the the scope of NSA mass surveillance was starting to be uncovered.[1][2]

[0] http://techcrunch.com/2015/01/11/leaked-palantir-doc-reveals...

[1] http://www.businessinsider.com/palantir-wikileaks-apology-20... [2] http://www.businessinsider.com/palantir-wikileaks-2011-2?op=...

> 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.

Militaries and intelligence agencies do much more than just make kill lists. And I sincerely doubt such lists are created without human supervision. But being mostly a pacifist I've never wanted to work with those projects, so I don't know the details.

disclaimer: I work at Palantir.

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...

Apparently they're not doing well. Good riddance. Palantir is creepy as hell. For all the snark about how startups and places like Apple are run like cults, like in The Circle, Palantir is the one outfit I know of where the stereotype is really true. And in the post-Snowden world where we know for sure just how all-devouring the US intelligence's community's desires are regarding our privacy, the last thing we need is Silicon Vally outfits helping them out. Assuming Karp can't right the ship again, I will not be sorry to see them go.

It's a great article but I agree with you: to heck with Palantir.

To hell with Palantir and the people who work for it.

I'd sooner serve fries than contribute to the shit they do.

"Spies Sans Frontières? How CIA-linked Palantir is gaining ground in the aid industry (and why some humanitarians are worried)"

https://www.irinnews.org/special-report/2016/03/07/spies-san...

Software like Palantir's is a classic example of a dual-use technology.

Sure it's creepy, but there's also something else I've been suspecting about all this surveillance for years: it doesn't really work.

Moral, ethical, and political problems aside is mass surveillance snake oil?

Is there really any credible evidence that mass surveillance has stopped any significant terror attacks? Can you really find needles in haystacks reliably without drowning in false positives?

It is possible that no significant terror attacks were even attempted against the US since 9/11. And if an attack was stopped they wouldn't necessarily want to publish how they stopped it. So it is impossible to really know if the surveillance does anything.

> It is possible that no significant terror attacks were even attempted against the US since 9/11

The TSA itself has literally admitted that there is no evidence of terrorist plots against aviation in the US[0].

That doesn't cover everything that could be possibly labeled 'terrorism', but given that piece of evidence, it does raise more than a little doubt about the prevalence of threats elsewhere.

[0] https://boingboing.net/2013/10/17/tsa-admits-terrorists-in-a....

Since when is the TSA an intelligence service?

They're mall cops.

>And if an attack was stopped they wouldn't necessarily want to publish how they stopped it. So it is impossible to really know if the surveillance does anything.

A trial, send a group to gitmo, a drone strike? If they could provide an example it would be plastered all over the place. Hell they even post youtube videos of their drone work - https://www.youtube.com/user/centcom/videos.

> So it is impossible to really know if the surveillance does anything.

We know that it has a chilling effect, and the likelihood of it actually performing the stated mission is really low. We've had several massive leaks, but the one place that evidence would have definitely shown up would have been the DOS cable leak - because it was an unfiltered dump. While SIGINT certainly shouldn't appear in anything below top-secret, there would be hints. The stratfor leak would have contained hints as well, due to all the candid emails discussing intel from gov sources (see access to OBL intel cache for an example). Also, the FBI would likely brag about more than their sorry list of terrorist entrapment cases. One can't make the claim that operational SIGINT is so special that it never leaks, because that is one of the classified emails found on Clinton's private server - emailed from a former government worker that shouldn't have had access in the first place.

What would you consider "significant"? There's been a number of terrorist attacks post 9/11 and I'd consider at least a couple pretty significant (San Bernardino, Fort Hood, Boston Marathon, Charleston Church). There's been a number of foiled plots too: see http://securitydata.newamerica.net/extremists/deadly-attacks... and http://securitydata.newamerica.net/extremists/nsa-analysis.h...

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 caught?

It would (will) be a very different movie, to say the least.

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 nuanced understanding of the intelligence failures that lead up to that event. They were a laundry list of things that the FBI (and others) had been proposing for years that they immediately dusted off and rammed through a congress terrified to say no.

If you want to roll back the surveillance state, you have to actually empower people carrying out anti-terrorism to intelligently correlate data gathered through lawful and privacy-respecting means to succeed in their mission so that lawmakers and voters trust their own safety to allow that to happen -- and prevent as many massive terrorist attacks as possible, because every one that succeeds will lead to more demands for more data and whatever it takes.

Palantir was specifically founded to give the authorities the right tools to solve anti-terrorism based on a focused dataset. The tools aren't designed around mass surveillance nor is anyone I know who works there interested in or motivated to solve that particular problem.

>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?

If you're opposed to a surveillance state, you should care deeply about preventing future terrorist attacks

You can't prevent them all, and I don't want you to try. More precisely, I don't want you to do what's necessary to prevent every likely form of attack.

Terrorist attacks in the USA happen when a few nutcases get lucky and succeed in wielding disproportionate force in an asymmetrical fashion. Only a ubiquitous Panopticon of a police state has a chance of stopping them all. In infosec terms, patching all of the vulnerabilities will require us all to live sandboxed lives far away from the metal, and that's not what this country was supposed to be about.

So, go find something else to do, please... or at least, someplace else to do it. Failing that, maybe name your company after something besides a powerful magical artifact that was captured and subverted by malicious forces. The real terror will begin when the next J. Edgar Hoover takes control of what you've built.

Bullshit. The authorities had all the data they needed to stop terrorist acts from 9/11 to the Boston Marathon bombing. They needed more brainpower, not more CPU power.

> “We’re looking to do transformational work with our customers,” Gavin Hood, Palantir’s chief of staff, said in an interview with BuzzFeed News. “Finding the right partner to do that transformational work takes a lot of care and a lot of attention.” He added, “There’s a lot of reasons why that doesn’t always work out.”

Sure it's hard to find suckers who will give you million of dollars for poorly defined return.

Theranos, Zenefits, palantir, what's the next unicorn deflation? Is it Uber or Spotify with their Palantir-esque spend a $1/make $0.90 strategies or is it an another company with less-reported vulnerabilities?

"Although the company is not profitable"... Excuse me but, you have to be fucking kidding me!!

Nope. Pretty much any nonpublic silicon valley companies aren't -- some public ones too. They are spending more money then they have, in the short term, in order to become bigger and more successful in the long term. That's why the venture ('building something') capital ('with money') industry exists.

Will Alden is shaping up to be one of the most connected journalists in startupland. First the Zenefits coverage, and now this, he's done a great job getting lots of great sources at some of the more interesting companies.

Credit to BuzzFeed for building up a pretty solid news team.

Buzzfeed was also respnsible for outing tennis fixing in The Tennis Racket using statistical analysis of players' matches.

Buzzfeed is shaping up to be the one of the best sources for investigative journalism. Who would have thought?

A lot of other places can't pull it off because investigative journalism doesn't pay all that well. But it looks like drawing a strict line between "clickbait" and "investigative" while serving both is actually a win for Buzzfeed. Clickbait makes money from "most people", investigative journalism encourages thought and shows off how good Buzzfeed can be.

Indeed. It looks like modern publications need not just a separation of church and state but also the bar across the street from the church to bring some foot traffic in the area.

AFAIK, that was always the plan. You need a huge audience to have the revenue to do great journalism online, and one of the few proven ways to build that audience is with clickbait and listicles.

So while it may seem like Buzzfeed has a split personality, it's very much intentional.

While I'm sure plenty of people might disagree with my sentiment, I feel Gawker had a shot at this back in the day when they had Neetzan Zimmerman writing all the "viral" articles and bringing in some majority percentage of the daily PVs. This then gave room for others to do some great/interesting journalism.

Buzzfeed can credibly do investigative journalism because despite all the clickbait, they maintain a good reputation. Gawker on the other hand is the Trump of journalism, absolutely shameless, willing to do and say anything for clicks. No amount of investigative journalism will ever allow them to free themselves of the hole they dug themselves into.

Yeah because for all the flak it gets, Buzzfeed's "trash" content is very benign. Cat GIFs might be annoying, but they are nowhere near harmful, unlike the kind of crap Gawker pulls off

Its not a crazy business model. My early stage company gets nearly all of its revenue from a product no one in engineering has touched in over a year, which funds some more intensive and risky product development. Similarly, DuoLingo has a foundation in teaching languages while building up a large supervised data set for machine translation. A lot of companies are basically fronts for getting average folks to either curate a label set or provide permission to access existing data. My old company provided a vague statement like, "Improve your experience, integrate your gmail account", providing quite literally nothing to customers except remove the button, yet provided us unlimited access to contact books and emails/attachments, many of which were very private and confidential. Not proud of that one, but thats the reality and for some people, very profitable.

Wow, that's nuts. My gmail account is the gateway to password resetting a bunch of different accounts, not to mention all the private correspondences. It would have to be quite the value proposition to make me give up the keys. And these idiots give it away for nothing at all?

Yes. We had numerous SSN and bank account details, basically anything you might find in a personal or professional email inbox. Obviously we weren't making use of any of this, but if we had one malicious engineer, things could have gone very poorly. One automated tool inadvertently posted an individual's SSN on the paid version of our site was very angry. The level of access was something we were always discouraged from discussing with sales folks and management always kept things very vague.

Holy... Okay, just went to https://security.google.com/settings/security/permissions and revoked permissions for a bunch of apps. Thanks for this story.

Can you actually give that kind of gmail access to third party apps? Luckily I don't but it's unnerving to think that doing so can have those ramifications.

I'm not clear exactly what that old company of yours did, and its putative value proposition for clients. Remove a button, where for what? Very curious!

No one would have thought.

That's why these comment threads are always the same. People remarking how surprised they are that something of this quality came from buzzfeed. I see this as a desperate need to separate their investigative brand from their click bait one.

Kind of amazing when you consider their first 5-6 years, actually.

Maybe there's hope for us all, is the way I look at it.

My theory is that when you're doing 90% cheap clickbait crap, you probably have a few dollars extra to spare and use it in paying good writers for actual good articles which will give you enough clout with people who actually cares about good articles.

I believe the New Yorker did a piece about building up their news department, which included hiring a couple of Pulitzer winners to the staff, but I don't seem to be able to find the article.

It might not have been the New Yorker, but rather the New York Times. Is this the article? http://www.nytimes.com/2013/10/22/business/media/buzzfeed-hi...

Found it. It was a third New York titled publication, nyrb:

http://www.nybooks.com/articles/2015/06/25/digital-journalis...

> If you have information or tips, you can contact this reporter over an encrypted chat service such as Telegram, Signal, or WhatsApp, at 310-617-1302. You can also send an encrypted email to will.alden@buzzfeed.com, using the PGP key found here.

This is the best "contact me if you have juicy insider info" slug that I have ever seen. He can take PGP-encrypted email!

And this is on Buzzfeed!!!

Buzzfeed's long-form journalism is pretty good/interesting/relevant, you just have to hold your nose long enough to get past the foul stench of click-bait.

The tennis racket article is very good. https://www.buzzfeed.com/heidiblake/the-tennis-racket?utm_te...

They seem to be putting quite a lot of effort into producing actual journalism lately, and I agree it's good (I also enjoyed the tennis piece mentioned).

Well that or use HN to curate for you, the extremely rare well written pieces are not worth frequenting Buzzfeed, Techcrunch etc for

That's how the economics of journalism has always worked. Going after people with a lot of money at stake will never, ever pay the bills and is a great way to rack up a lot of legal fees. So in the old days you surrounded the real news with classifieds, comic strips and movie listings, and these days you surround it with cute animal videos.

Even Gawker writers take PGP-encrypted email these days.

Could a publicly traded newspaper company legally allow its employees to use end-to-end encryption? I would guess not, given the email retention requirements in Sarbanes-Oxley.

Some companies even block webmail providers to ensure only the centrally-archived corporate mail server can be used for company business.

Remember that emails can be encrypted to N recipients. So long as one of the recipients is a public key corresponding to the legal team at your company, it should be fine, as they can decrypt and read anything they would need to. They would also keep their private key in high security offline storage. You'd just need to configure your PGP mail plugins to always also include the legal team's key as a recipient.

If I had to make a choice to contact the author, I would go for Signal. You can make a lot less cryptographic mistakes by using it, and it doesn't backup all the conversations to Google Drive like WhatsApp or store them in the cloud like Telegram.

the big problem that stuck out for me is that it is an unsigned pgp key hosted on another site. Which means someone could easily replace the key with something else and man in the middle the communication. At least have a couple trust worthy public sources sign your key if you going to asking people to communicate with you about very sensitive information

This article was way, way better sourced than I thought it'd be. Especially for Buzzfeed of all places.

This is a trend I've noticed. BuzzFeed has had a number of excellent longform pieces recently.

They're diversifying their offering. The clickbait stuff won't change, but they're looking to build credibility with those longform pieces you mention to shed their poor reputation.

for me, buzzfeed is like hyundai. Hyundai had the reputation prob anytime before 8ish years ago of being total unreliable shit. However, they made a massive set of enigineering improvements, created better styling and started standing behind their products with the best warranty in the industry, 100k miles. They even created a luxury line under the hyundai brand and started offering more expensive cars. During hte period, their stock started actually doing well[0].

It has since, gone down a bit and I won't say it is 100% the point I will make next,but it factors in. Buzzfeed is a lot like Hyundai. While Hyundai killed in the low-cost market, no middle class consumer wanted a hyundai. Their brand had negative value.

Just as quality longform content is not read by the same people who read buzzfeed. So they(bf) will not adopt users, just as hyundai slipped a bit after the marketing relaxed and consumers remembered that the "image" conveyed by hyundai was low-income/cheap. While they had erased the real problem, shitty engineering, the image was just as important.

So buzzfeed should rebrand or meta-structure itself if, at least in my non-expert off the cuff opinion, it wants to ever sell content. Obviously, ancilliary advertisements are will be dead very soon, but buzzfeed is really good at writing advertising copy, they call it journalism i think.

However, as stated above and evidenced by this very article they are breaking some good stories, and are funded by a16z. They have quality journalists who have access and are equipped & talented to write great articles. I just think that most of the people who would value that article, would never read it from buzzfeed. buzzfeed should channel out services like vox and rebrand.

tl;dr know your customer

[0]http://finance.yahoo.com/echarts?s=005380.KS+Interactive#sym...

Hyundai's transformation happened much earlier than 8 years ago. By the time I bought my first car in 2002, they already had a decent reputation and a 100k warranty.

From a linked article:

"The primary payday for the best engineers is that you get to work with the best engineers,"

Funny how that's not the same for execs!

There are a lot of similarities to Theranos.

SAP-like scam. Highly refined, Oxford-educated sales and execs, selling with high-status deceptive techniques grossly overpriced, lovest quality outdated by a decade crap, for support and maintenance of which they are billing their victims for each hour spend by expensive suit wearing nonsense talking consultants.

BTW, the guys who went there for sweatshop positions are those who barely managed to graduate. But in sales there are top tier liberal arts guys, with refined speech and behavior.

These kind of enterprises is a classic social pyramid, modeled after an organized religion, especially Catholic Church, with exactly the same deceptions, dogmas and loyalty for lower ranks.

> 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.

Eventually, one finds oneself with a few guys in expesive suits who are sitting in the car right on your lap, totally obstructing the view, but telling you when to accelerate, stop or turn, claiming that this is a much safer and the most effective way to drive a car, so they bill you for each kilometer and each turn taken.

They, moreover, have replaced your car with their own ancient vehicle, to which there is not a single mechanic in a 1000 miles and the only spare parts available from their own garage, with the replacing procedure which requires to remove everything from the salon and from under the hood and then to reassemble it all back, which takes a few days and for which you are billed for each hour of each of twenty certified technicians, half of them flown from the head office at your expense.

And, please, don't tell me that consulting companies are different.

Btw, really smart top grads would never even come close to these scam shops, like sane people would avoid Ponzi schemes or a telemarketing crap with a hidden charges. Being fooled by meaningless titles and certificates instead of working with a cutting edge technologies and obtaining skills which could be sold later.. I can't imagine top grads being so clueless.

> 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.

I'm curious if any positions require pre employment drug tests, random urinalysis, or security clearances. What a miserable life for a 20 something in the private sector!

You are aware there are many people out there with technical skills who choose not to use drugs for a multitude of reasons? For these people, a security clearance can add tons of job possibilities, a higher salary, and greater job security, with the only tradeoff of being more "miserable" according to the occasional HN poster.

Clearly some jobs at the org require security clearances, but most don't, and access to career growth isn't gated on that at all.

Look up the forbes article from 2013 about the CEO Alex Karp (can't link easily due to adblocker BS) entitled "How A 'Deviant' Philosopher Built Palantir", and tell me again with a straight face that you'd be worried about this.

Reply on news.ycombinator.com