> Average TC of 900K
> Several unranked billionaires
it makes even OpenAI / Meta ML SWEs look underpaid
SWE who have deep subject matter experience are super valuable to these firms. Folks who understand how to write low latency code, FPGA work and other stuff like that. But the real money is in figuring how "how and what" to trade. Once that's done, the SWEs can bang out the code.
JS pays their new grads similarly, just a tiny bit lower.
New grad SWE is 400K
https://www.levels.fyi/companies/jane-street/salaries/softwa...
And levels.fyi not very accurate because 2nd year bonus is much larger than 1st year bonus in offer letter.
Quants/traders can hit $1M with 5 YOE not too difficult. Portfolio managers (similar to EM in tech) definitely $1M, sometimes $10M.
900k is average including janitors and HR.
Median for SWE most definitely over $1 million. Turnover of 6%? No way people stick around “only” making 500-600k new grad comp. Check levels.fyi
> Also, some high-level SWE do make 900K at Google and others
And some high-level SWEs at JS make 9 figures, not 900k.
Tether made 4.5bn net income in Q1 2024 w/ probably <50 employees.
Because of this, getting a SWE role that is assigned to a trading team is highly desired.
Superfast arbitrage, which is Jane Street's main thing as far as I understand, doesn't really produce anything of value to humanity. Yeah, I guess the contributions to OCaml are worth something, but maybe not 10 billion?
They've never been at the forefront of latency games, like Jane Street isn't the firm sweeping equity markets since they have the fastest radio network out of CME (dubious value) or getting their quotes first-in-line every time.
The characteristics of short prediction horizon strategies like in HFT give you an amazing sharpe, but you're predicting small returns.
To scale returns you need to invest more, which means higher market impact, which means longer prediction horizon, on larger price moves.
it's the operating system, some people might think it's a tax on everything, some people might think it provides the foundation to produce everything of value.
similarly, Google is the high-order-bit in the information or content economy, the creators get underpaid, the people who do ad optimization get overpaid.
no financial markets -> no IPOs -> no VC -> no Google and Silicon Valley as we know it.
the closer you are to the money and the transactions, and the high-order bit, the better the opportunities to redirect and organize to your advantage, and the more you get paid.
The entire tech industry has fewer billionaires than quant finance.
If you are ambitious, tech SWE is a bad deal.
And I would bet that tech SWE has a far higher probability of launching you into $10M+ range than quant finance.
I don’t know that a few more single digit billionaires (if that is even true) puts quant finance ahead of tech SWE in terms of potential rewards for the “ambitious”.
Also, quality of life is far better for tech SWE.
I've rarely met anyone who is a quant and doesn't have a PhD. That already narrows down the field quite a bit.
$1B net worth? Google has more SWEs than the entire quant finance industry globally. Yet Google could only produce what, 4 billionaires?
$10M TC at Google? That’s at most 500 VPs out of 150,000 employees. That’s 0.3%. Not good odds. More than 7 people (0.3% x 2500) at Jane Street are making $10M+…
Same logic applies to $1M TC. Tech has okay chances, but quant is simply multiples better. Only if you got lucky at Tesla or NVIDIA. Otherwise your L5-E5 comp maxes at 600-800k including RSU appreciation.
Then for junior/new grad, quant firms win again. 300-500k quant SWE offers vs 200-300k in tech.
Hell, even for summer internships, quant trading firms pay SWE interns $20-25k/month. Also on levels.fyi
[0] https://www.youtube.com/user/standupmaths [1] https://signalsandthreads.com/ [2] https://www.janestreet.com/puzzles/current-puzzle/
My impression with them in general was that they were willing to do lots of things that did not "conventionally" make sense at their size, and those things paid off. The internship program, for example, was relatively large in comparison to the size of the company at the time (≈50 interns?) and had lots of structure (events/talks/classes/group projects/etc) like you would expect from a big tech company, not from a 300-person firm.
They were also willing to build a lot of tools in-house like their own build system[1], their own code reviews system[2], etc. Most people would see this as wasteful NIH but I'm convinced it was a net benefit for them—they managed to be so productive in an absolute sense and especially on a per-engineer basis because they were willing to build so much themselves, not despite it. I'm sure the same thing applied to their recruiting efforts.
The biggest thing I took away from my internship was how much conventional wisdom in the software world was not necessary or true.
[1]: First Jenga, now Dune
[2]: Here's a neat talk on how they do code review at Jane Street: https://www.janestreet.com/tech-talks/janestreet-code-review...
It paid off for them, but I would say it's a risky move in general - it is very easy to get sidetracked with failed projects that have nothing to do with the business. There are many examples - Uber's internal chat system, etc.
Others tools were more a necessity, eg the OCaml build story was not great back then. Investing in OCaml itself was a calculated risk.
Also I wonder how things would be different if they were starting today, where there are more mature tools in the space.
Do you remember any specifics?
I’m also curious how their tooling made them so much more productive. Were the tools just really well designed, or did they integrate perfectly with each other?
And yeah, jane street is a pretty compelling demo that A) NIH syndrome is fine if you're good at writing software and B) it doesn't really matter that much if you use a mature language or some uncommon immature language
Because it is. What is the point of reinventing these wheels when gazillions of man hours have already been invested on open source tools that can do it better and cheaper?
Moral qualms aside, it must be a fascinating place to work.
> At the end of 2023. Jane Street employed 2,631 people, so that equates to almost $4mn of net revenue per head on average. In adjusted EBITDA, it comes to $2.83mn per employee (or nearly $22mn for each of the 482 traders actual traders at Jane Street.)
And regarding compensation:
> Given Jane Street’s disclosed compensation and benefits of $2.4bn last year, this works out to over $900k for each employee on average.
When I first heard of Jane Street, it sounded like Yaron Minsky was going around to MIT and such, high-touch, trying to hire just a few people. And later, things like this blog: https://blog.janestreet.com/author/yminsky/
The only negative thing I recall hearing is Jane Street alumni responsible for the infamous FTX and Alameda Research. I don't know whether the individuals were already fully into hopped-up sociopathic/narcissistic thinking in college, or whether their internship and employment experience contributed.
the bigger the vol, the more imma do
Is learning C++ a must?
For quants (progression towards a trader / portfolio manager), Python / Matlab / R is enough.
> About 80 per cent of the company's capital comes from employee equity, which has swelled to $21.3bn at the end of 2023
o.O
> A good mate who runs and wrecks expensive cars for hobby once reminded me that course plotting (i.e. research) and braking (i.e. risk management) are the two sine qua non contributors to a successful race.
Along with a reminder that disasters happen when businesses forget this (Boeing!)
Meanwhile I'd love to know what their edge is... It's probably more than OCaml, although... ;)
They are not making their money market making equity etfs afaik, mostly bonds. A slightly different game with a much higher barrier to entry etc
If you want to hire the best and willing to pay that is no longer a concern
They use OCaml because it is explainable. Which it is. They use it like a non-lazy version of Haskell -- side effects are used rarely if ever. So there's no nonlocal behavior in the code, which makes it easy to reason about. And that kinda matters when a lot of money is at stake.
Incidentally, Standard Chartered has their own compiler for Haskell, without the laziness. The group is led by the guy famous for Cayenne (the first dependently typed Haskell dialect).
Most companies don't even know that these are programming languages. Also don't assume people interested in these languages are "the best". Regarding JS, this is all hearsay, but I think nowadays they hire good profiles (competitive programmers / olympiad winner / top graduates from prestigious schools), ask them hard leetcode questions, and teach them OCaml (which anybody can learn, not particularly hard. Their talent pool isn't restricted to the OCaml community, as it used to be when they were less famous, except for niche use case (compiler work and so on...)
And one of the answers was Jane St. Apparently they produce great engineers.
Yes, I suppose this is all something to get all starry-eyed over, Jane Street encroaching on Citadel Securities, the two of whom control 30% of the US equity market volume.
I see it another way. I see people's hard earned money being siphoned by enormous financially-engineered vacuums, never to be seen again. And not just in the US, globally. This won't stop at 30% of the US equity market. It won't stop until the music stops and the last chair breaks. Which may or may not be soon. It will certainly be coming at some point.
Five times the London Stock Exchange’s entire trading volumes in 2023 in just your ETF arm? Sure ... this sounds like reasonable growth ...
> This is why some people argue that APs like Jane Street have become systemically important.
Oh, you don't say!
> About 80 per cent of the company’s capital comes from employee equity
That's adorable. They're like a little mom-and-pop shop ... except not anything like that.
Which means they're rapidly coming to a position which will easily allow them to game the system (what used to be known as "cornering the market").
I even wonder if their system has already learned cornering by itself via stochastic gradient descent.
These are flows, not ownership.
Also, taking Citadel as an example, a vast portion of these flows are not their own, but rather thirs parties offloading their flows to them, so they have the underlying best execution guarantee to provide.
An other way to look at it is that investors are moving away from traditional brokers to execute their flows, because these HFT firms have become so good. So instead, investors offload their flows to HFTs acting as DMM instead.
Their effort definitely paid off though. Quants like JS, Citadel, or Jump hire some of the brightest students from Berkeley and other top CS schools.
That's a lot of money for someone who is essentially a middle-man. What a grift, about 13 dollars for each average American lost out to them a quarter.
Second, it's not at all obvious that market-making is a zero-sum game, where profits to the market maker are "lost out" to someone else. Market-making profits from the bid/ask spread, but a new market maker tends to reduce that spread. Ordinary people tend to invest as price-takers, and thus they benefit from a reduced bid/ask spread.
> We operate as a functionally-organized structure consisting of various management and risk committees. Each committee is responsible for directing the overall strategy of the firm and for emphasizing the importance of risk management to our operations. Each of our trading desks and business units is run by equity unit holders who take an active role in managing our day-to-day operations [...]. Our management structure allows for effective cross-departmental communication [...].
Yes, that org structure full of hierarchical business units with distinct responsibilities and committees and subcommittees and overseers and risk management totally screams "anarchy"...
I'd be interested to see if the Pareto distribution holds here as well, namely that 1% of employees (26) hold half the wealth ($10b).
> The real money is at the top. The bond prospectus reveals that Jane Street has 40 “equity unit holders on a full-time basis and in good standing”, with an average tenure of 16 years. Among those there will be at least a handful of billionaires, even if no Jane Streeter appears on any rich lists.
Sounds like any other partnership. A few people at the top are providing the equity and getting a profit share, and the thousands under them are getting salaries.
The difference is that these thousands also get to invest in Jane Street, which seems a pretty profitable investment (70% margins, etc).
Jane Street shares and profits are proportional to the capital you invest/accumulate.
commodities trading houses tend to follow this model too though that is changing a bit.
i remember having this discussion with a friend after he sent me a richard wolff video. nothing about our system stops coops from flourishing. one of my favorite retailers, REI, is a member-owned co-op. publix, the beloved florida grocer, is employee-owned.
Finance tends to pay its workers better than shareholders—most banks’ trading and IB groups pay out more than 50% of profits to workers.
Where it goes wrong is when the regulator fails to stop foul play…
Yes, Jane Street uses OCaml, they have no reason to stop using OCaml, but may very well have been just as successful using another language. It's hard to tell, when we don't know the full circumstances of why they went with OCaml initially.
> but they could have been just as successful using another language.
His thesis is that finding one really good engineer in said language was 1 in 10 (10 interviews to find 1 really good engineer) whereas in more commonly used languages like Java, JavaScript, etc., it might be 1 in 100
[Edit] https://www.juxt.pro/blog/clojure-in-griffin/
If we had picked Python, it’s very boring and reliable, and the same could be said of Java. But you’re picking the lowest common denominator. I would say high performers, and the best programmers are often people that will only work in niche languages.
The problem is, there are good Java programmers, but there are also thousands of terrible Java programmers. If you pick the right niche, it’s easier to find the high-end talent. I think Paul Graham also made a very strong case that in a startup, you should be using the most powerful language you can, and that is Clojure.
I don't claim to be a rockstar developer or anything close. But my capabilities and efficiency as a developer are tightly coupled to the tech stack I use (not just language).
I moved from a job where I chose my own tech stack that I iterated over several years to one where I'm forced to use (IMHO) tools that are poorly suited to my work, and I'd say the quality and volume of my work has dropped by at least 10x.
So I think it's both. You need smart people, but they also need to be using the right tools for the job.
However, I think if you are a company doing something boring and that can only pay average, then having an interesting tech stack (including a nice language), hiring globally and having good benefits might give you a venue to compete for talent. You'll need some kind of strategy.
The paradox being: developer familiar with programming languages of level of power N doesn't recognize that languages of level N+ are better (more powerful expressively), only that N- are lesser.
These days, starting or running a financial business with less popular languages is, well, less popular.
can you expand on this? I have zero idea of what Jane street actually does and how they actually make money. (someone wrote that they have ~450 traders. trading what? equity? stocks? dark pools? PE? are they market makers? are they offering services to institution types?)
also what does "people's hard earned money" mean? you mean that Jane Street takes away their 401k or ... ?
We could essentially close them down if we moved all trading to say 1 hour a day.
We could essentially close them down if we moved all trading to say 1 hour a day.
Though most retail traders are upset they can only trade 7.5hrs a day, they want to trade 24/7 instead. They seem to WANT the liquidity and are apparently willing to pay for it.
Me, I'm fine with trading for only an hour a day, but I get not everyone is as lackadaisical as me.
Unless you have a better mouse trap to solve the liquidity problem, this is the best we have for now.
I actually know a little about this space. You know what the easiest response is that is always the answer?
"We provide liquidity".
Sounds important, most people don't get it, it works.
But it's not like the market is going to grind to a halt if Jane Street disappears overnight. Of course, people will say that. Not people telling you the truth.
I actually considered "but, but, THE LIQUIDITY!" in my answer but I felt there was sufficient snark.
That's funny, people say the same thing about ticket scalpers.
No. Flow data is quite expensive, and most providers will only provide bucketed data unless you contribute with your own flows.
OP obviously has no idea what he's talking about but hey, what wouldnt you do for some internet points.
Envious bullshit! The reason they are so profitable is that believe it or not they are replacing earlier operators who were less efficient and taking more transactions costs out of the system before. To be anti-Jane Street is the same as being pro-Big Bank of the past, that was taking more money out of the economy doing a worse job!
Unless you know the history of global financial markets and lived it then you don’t understand. For example in the late 80s the CBOT treasury bond bit had over 1000 traders in that pit. They were all making money and quite a few a huge amount. And there was many more people supporting those traders of the floor. That was just a single futures bond contract! With Jane Street we have 2,631 employees doing the equivalent job globally and for less total cost of at least 80,000 employees in late 80s.
The profits per employee are higher obviously but that’s the effect of technology and productivity but the total price being charged to the economy as a whole is much lower. This trend will continue and I would not be surprised in 10 years that a company of 200 people will provide the entire function of those 2600 today, and probably the profits per employee will be $10M per person. But that’s what we want, is a good thing not bad, portraying otherwise is just Envy.
No, no ... it's not.
First of all I don't care one iota about Jane Street or how much money Jane Street employees make.
The problem is that this money has to come from somewhere. And while you say it's due to the entire system becoming more efficient, that's not the full story. That $1 billion Indian options play came at the expense of Indian retail investors.
So yes, there is a siphoning of money out of the hands of the retail investors and into the vast pool growing under Jane Street and its employees, much as it has done under others like Citadel.
Who is that good for, except Jane Street?
If you look the per capita income from where it's leaving to where it's going, that's where I see a growing problem.
There may be a sucker born every minute, but that doesn't mean that this is all just about making markets more efficient.
And, of course, none of this would have been possible without the additional "liquidity" provided ... the clearly essential contribution that makes it all justified. "We provide liquidity!". Ok ... ?
There's some efficiency improvements too, but ... eh.
And instead of (metric) tons, I want megagrams. No kilotons; gigagrams.
I've had people get angry when I tried to use Mm (mega metre) or ask why using some currencies symbols before the number when all symbols for all other units are used after.
B) if you have work experience, they filter on where you worked because the average google engineer / HFT engineer is probably better than your average engineer who works at missouri national bank.
C) if you've done something great that everyone knows about (like being the author of popular libraries, inventing tools that people use), then you can most likely bypass filters A and B if you can get in touch with a human recruiter (shouldn't be too hard).
For prestigious smaller companies which gets a lot of applicants, there is no easy way to reduce the pool without doing A and B. It is unfair and what college you go to might depend on circumstances beyond your control but if you have high ability regardless of what school you went to, you would eventually do B or C and get into the prestigious company.
If you don't have high ability and will never do either B or C, then they did the right thing filtering out your resume.
In the real world, it doesn't make sense for a company to discriminate against a candidate with higher ability just because they went to the wrong school (unless the company makes money from appearances like law firms, consulting firms or is corrupt and entrenched).
That being said, if you went to a community college but also had 10+ years experience at one of their competitors (eg: DE Shaw, 2Sig, Citadel, etc), you'd almost be guaranteed an interview. But once again, thats another very small set of applicants.
It wasn't a "logical way", it was an "easy way". That's classist at the very least.
The question is what other attributes you "accidentally" filter by if you filter by school prestige. Even if you do bias your pool to individually better candidates on average, you may bias it in a lot of other ways.
If the purpose of the filtering is to reduce the pool to a size where humans with common sense can make sensible calls, then that CC graduate with 10+ years of competitor experience would likely be filtered out before common sense could suggest he might be good anyway.
they get many applicants for a job. So you have to filter those applications. One way is to recruit from known places where they have already got people from as they know it worked. Typically those universities also filtered on how good you were and so the average quality is higher from those.
Obviously there will be exceptions but not work spending time to find them. Recruitment costs a lot in time and money.
I have applied about a dozen times since then after getting my degree from WGU and they haven’t gotten back to me; I almost wonder if not putting any college experience on the resume was a blind spot for their filter.
You can't claim to have diversity if everyone did the same courses in the same schools.
> the organization of society on the basis of voluntary cooperation, without political institutions or hierarchical government; anarchism.
Because people seem to think it's "politics" and "government" that are the problem. It's not. It's people. It's always been people. It will always be people. People are the problem. Failure to acknowledge that people can and will be shitty is going to doom your system before it starts.
Sounds like a company.
Think when trying to translate a society to a corporation, then each of the terms must also be translated.
For a 'society' there is a government. And politics is dealing with the government policies, how we come up with rules.
But, if you want to translate to a corporation, you can't just say it doesn't have 'government and politics' because the corporation isn't dealing with the larger society. You need to also scale down the terms,
The CEO and the Hierarchy of bosses, are the 'Government' and 'political institutions' are the departments like HR, Marketing, and between them there are 'politics'.
There is no real 'anarchy' like individuals having freedom to 'just work on what they feel like'.
The things that made them great traders also made them much riskier choices to lead a larger company. To them, compliance is an obstacle in the way of profitable trades.
How do you quantify, or even loosely, determine that?
What about ancillary workers who might not add any significant value to the (e.g. office janitor or HR) but supposing there’s a serious janitor or HR shortage then the org will still have to pay enough to attract someone, but what they pay is outside of their control and unrelated to the actual value to the company. And even in a worker-owned cooperative there’s still going to be in-groups and out-groups, and the in-group is incentivised to pay the out-group as little as possible as to maximise their own returns.
Artists set too low of prices because they don’t want to be thought of as greedy. Scalpers expose the true market rate of the goods.
Removes any confusion for european and american collaboration.
Got old roots dating back to cockney trading slang: "The Old Lady just bought half a yard of cable"
So there was confusion, even in the English-speaking world. Was. If I understand correctly, England has adopted 10^9, and so now there is no ambiguity.
HFT jobs are high value / high comp jobs, so Jane Street and their cohorts put a lot of effort into recruiting. This includes their resume intake system, internships with desirable colleges and professional outreach.
I think fundamentally a merge-based workflow, which git(hub) encourages, is kind of inimical to stacked PRs compared to rebase-based workflows.
actually, quite a good idea, regardless which sector!
But even taken at face-value, Bloomberg's annual fee is comparable to the yearly cost of an undergraduate degree - so it might very well be a fair deal.
20% of 20% of 20% (or 0.8%) of people will hold 80% of 80% of 80% (or 51.2%) of the wealth.
In binary, most significant(left most) bit is twice the magnitude of the previous bit. So that bit (or person) would contribute and get compensated accordingly.
Would also be similar to segment trees with each leaf node having a value of 1 with parents recursively summing the children?
If the data consists of electrical potentials between 0.0000001 and 0.0002 V, it’s way easier to think about it as microvolts, especially if any other is in a complementary scale.
speculators can then sell/buy to/from them and if they are doing it well, they make a profit. (and help narrow the spread.) sure, great, they even inject some liquidity. but why do we want a narrow spread? it only helps people who don't know which side of the trade they would rather be on, no?
The spread also compounds quickly. A wide spread indicates low volume and a sparse order book. This means if you need to offload even say 100 shares, you can single-handedly as a retail investor, widen the spread yourself and take an even larger loss.
Liquidity injects supply/demand and its always a good thing for retail investors. It goes past the stock market as well, spreads are why pawn shops, thrift stores, and even eBay can be profitable in certain goods and with other goods, not so much.
You can absolutely set limit orders, but depending on the limit, other people may or may not be willing to trade with you. Your order can sit all day and never be accepted.
If I want to sell AAPL, and set the limit oder to $1/share, buyers would take that trade instantly. But if I set the limit order to $10k/share, nobody would buy from me.
The exact same thing happens with the buy side.
It's like haggling with the vendor on the street corner or negotiating with your car dealer.
Think of it like a liquidity tax. Jane Street and other market makers(Citadel, etc) compete over that liquidity tax. The more often a fund, ETF or stock is traded, the lower the liquidity tax.
If XYZ Corp is very illiquid and only trades a few shares a month, then Jane Street will charge you a lot of tax to allow you to trade it instantly, because they have a harder time knowing what they can get for it a month from now when someone actually wants the trade.
You can see the wide range of bid/ask spreads for ETF's here: https://www.etf.com/sections/news/etfs-highest-lowest-tradin...
Notice something like SPY(S&P 500 fund) is basically free to trade, but something like EEH costs you a very pretty penny to instantly trade. For EEH, you might be better off re-issuing limit orders and just hoping someone comes along that wants that fund. If you let Jane Street or other market makers handle the trade, they will charge you more than it's worth to trade.
The only 'commonly' used unit of measure larger than a kilometer is a light year, at least that I can think of. Maybe a Astronomical Unit, but not 'common' I suppose.
But it shouldn't be common.
I'm all ears if you have a better system that is more efficient. I'd also add that I went to a State college and my resume would have been tossed if I applied as a starting engineer too.
But not socio-economically.
I know that fancy colleges like to say that they strive to base admittance on performance, and I believe that’s true to an extent, but it certainly seems like they make an exception for students who come from a lot of money. I’ve mentioned this before but think about any politician that you think is un utter moron, and there’s still a reasonably good chance that they went to an Ivy League because they come from a rich family.
And it annoys me that people act like Ivy League schools are the only ones that accept based on performance. I applied to a bunch of state schools as a teenager and I got declined by a bunch of them. We can wax philosophical about why I was declined but I doubt it was personal, just that they probably didn’t think I would perform well based on some metric.
I'd say that Jane Street would seem to be less hierarchical than others and thus more closely aligned to an anarchist collaborative.
(It's all relative, I suppose. Bernie is a "leftist" in the US and would probably be a centrist in Scandinavia)
Was it an intentional rhetorical device to brush past the important distinction?
Yes, there will always be more powerful cliques. This is Jo Freeman's The Tyranny of Structurelessness. But the point is specifically whether those cliques and hierarchies are encoded into formal committees or whether they remain fluid based on the desires of the group at any given moment.
If we avoid codifying the hierarchies, we get better distribution of information and resources, people are held responsible to those who delegated tasks to them, and responsibility can be rotated more freely.
What really happens is that important information and resources get hoarded because that lets you control things. The only people "held responsible" are those you can convince others to punish arbitrarily. You'll create an entire shadow government the majority of the group has no knowledge of.
And showing how it worked within a group of 4 to 5 people is kind of useless. Lots of things work on small scales. The minute you scale to the point where you can't keep track of everyone, you're going to need an actual system in place to deal with everyone.
I've been meaning to learn C++, but I always find it intimidating.
Which is probably what you'd want in a high-performance firm, less everyone look at the absent top level extractors and it turn into a law firm.
Fortunately it doesn't take that much to get top talent, because so many other companies underpay. Jane Street only has to pay out a small fraction of their PnL and doesn't even need to have a non-compete.
2. Building something for yourself is qualitatively different than building something for somebody else. (I've heard this described as "situated software"[1].) Both the results and the process are different.
3. Building something yourself lets you become an expert in the domain and the tool you're building, often faster and deeper than using somebody else's system. It's a way to build up tacit knowledge and institutional capital as much as (or even more than) software.
4. More often than people realize, building something yourself ends up simply faster than first learning and then wrestling an existing tool into the exact shape you need. I've seen a lot of teams waste way more time trying to get some existing thing working than they would have spent building their own thing.
Obviously it isn't always true that building your own version of something makes sense, and nobody is going to be building everything from scratch... but it makes sense far more often than conventional wisdom dictates.
[1]: Term originally coined by Clay Shirky, but I can't find the original essay online, but looks like Gwern hosts a copy: https://gwern.net/doc/technology/2004-03-30-shirky-situateds...
Moreover, one thing NIH syndrome handwavers miss is that developing a custom situation-specific solution allows you to expand your knowledge in that discipline, create a more tailored solution, and avoid supply chain vulnerabilities that can come with using third-party packages. In my experience a handbuilt solution to a problem is going to be much more efficient in most cases than OSS out there, except in spot instances where the scale of the task is not able to be achieved in a small amount of code (e.g. a versatile graphing library), although these are /very/ few and far between.
> 1. You can do it better, with better taste
You can, or you can just work around the existing tooling, which is quicker and assuming you choose wisely, has a wealth of googleable documentation. Its humbling, because you need to learn through _using_ rather than creating. It feels less productive, even though you're 90% of the way there with an existing tool
> 3. Building something yourself lets you become an expert in the domain and the tool you're building,
It might do, or it also might just drown you in complexities of the domain that you're building the tool for. Plus the assumptions you make when first designing the tool tend to be disproven as you learn more. Its good to question tools, but not blindly re-write them without studying _why_ they do that thing in that weird way.
> 4. More often than people realize, building something yourself ends up simply faster than first learning
Which means you are very likely make the same elementary mistakes as the previous generation tools. Plus its _always_ slower to start. mainly because naive re-writes have naive bugs. But you just don;t know it yet.
At the legions of VFX companies I've worked at, the number of people who look at the asset management system and go "Oh thats not hard, lets just re-write that to be x" is too damn high. 6 engineers and a year later, they still have a broken system, but in new and interesting ways.
Am I saying that you _shouldn't_ make tooling/software or custom things? no. I'm saying that you should really save yourself for something that critical to the company.
All projects have a limited number of innovation tokens. The more you use the slower your project will go. You should really only look to use 2 innovation tokens max.
Sometimes, the answer is to dive in an seek to understand it. Now, that may end up meaning that you fork the tool or contribute back to it. It may mean that you choose to build anyway, and it may mean that you come back understanding it. However, if a library is not working for you, the slowest answer is often the one where you fiddle with it until it's working without taking the time to dive in.
Since the best software is written by the best programmers, those of us interested in the best software, and have the capability to make such software, should be free to do NIH on an as needed basis.
As an example, say if you're a small cloud vendor, you should probably write your own machine OS imaging automation, but probably not your own chat client.
A few factors not normally taken into consideration;
- Tooling is a great place to practice transferable skills on a practical problem. This always annoyed me with the XKCD cartoon 1205 which didn’t take into account improvement in optimization ability from practice in optimization.
- I’ve had a few external library dependency rug pulls, open source and proprietary. It never happens at a convenient time and is a total pain in the ass. Contractual agreements won’t save you unless you’re ready to sue for breach of contract and even that is no real solution. Perhaps getting code in escrow could be an alternative but I’ve never seen that work out either.
- External quality is mixed and getting worse. I’m careful with my own code so most of the bugs I hit is in other people’s code. If it’s a small library it can easily be less work for me to bring stuff in-house than to debug someone else's crap.
Honestly I wish I didn’t have to bring so much in-house, it would have saved me a huge amount of work. But we don’t have an efficient market with clearly defined standard of goods where we could treat software as a commodity and I’m not sure if we’ll ever get that.
Letting your development team control and build their own tools is a good idea, if they are able to. You probably don't want to do it in Java or C#, it'll take too much time, too many people, and be too unreliable.
Hey, it's not massively relevant to your story, but this cracked a smile for me.
CSFB was a big Sun shop, up to and past the time they mounted a "SunSet" initiative around 1998. Scott McNealy came and gave talks to the engineers.
In CSFP, the derivatives arm, where the quants worked, there was more of a push to get onto Excel, build a relationship with Microsoft, which included getting onto the "Cool" program which turned into .NET.
You got the direction the wrong way around.
Ask to introduce OCaml at any company. The first thing is they're worried about is not being able to find people. To which the answer is to train them. But then they're worried some people just won't be able to learn ocaml quickly enough. To which my answer is hire people who are good enough to learn it quickly. But then they get worried those people are expensive. Ultimately it comes down to not being willing to pay.
And the equity holders would get diluted a bit when new employees are offered equity, but looking at the rate of profitability, each new employee more than earn their share in equity, even at the high end. Therefore, it is in fact, in the existing equity holder's interest to get diluted a bit to hire these employees, who would produce way more value (and thus increase the total value) compared to the loss in dilution!
In my books share ownership can be unequal, so long as the bottom rung still have a few shares each.
So many poor policies originate from Envy and poor reasoning not grounded in logic and understanding of free markets and economics. This would be yet another classic example of that.
> It would be technically negative in every measurement possible.
Since you believe this so strongly, can you provide academic sources for this perspective?
Maybe at the very top when you have more money than you’ll ever need, but for the vast majority of people jobs aren’t truly voluntary.
Yes, jobs are voluntary. The vast majority of people could choose not to have a job and not to entertain certain luxuries in life like having a roof over one's head or food on the table. Some even (voluntarily!) choose not to have a roof over their head because they don't like the idea of having a job.
Taking job A over job B could be described as voluntary but saying that you could just be homeless and scrounge food, especially in locations where the local government makes this illegal, so working is voluntary is just wrong
Food and a roof are basics, not a luxury. The fact that a significant number of people in this world don’t have those things speaks more to our failure to provide.
Do better.
>20% return is quite easy to justify. You're also looking at a 50%-200% annual bonus, mostly leaning to the higher end of the range.
Its a very different world!
Surely this is a HN culture bubble? Very few people can borrow tens of thousands of dollars from family and friends to lend to a hedge fund. Not only would I be refused, I'd damage friendships by exposing the moral vacuum at my core.
So they're incentivized to allow as many of their employees to invest as possible.
I remember the first time I got those units out of gnu units on the command line scratching my head wondering why I'd never used them before and why it was using them at the same time.
atto-lightyear = 1e-18 lightyear
lightyear = 3.156e+7 lightsecond
Hence, atto-lightyear = 3.156e-11 lightsecond = 31.56 pico-lightsecond
Anyone aged around 60 or older would have been learning these numbers before 1974, and that's a significant overlap with FT's audience. You are dismissing this as if it's Middle English. It's an entirely reasonable explanation, at least one that shouldn't be dismissed.
From the writing of the era, it's clear this new definition for million was not popular, and many chose to continue using the "British meaning." So it was probably in colloquial use for quite a while, and the transitional term "1 thousand million" became the proper style.
The FT has a big international audience. As a German reader where a "Billion" is still 10^12 it does sometimes trip me up a little. So I at least find it useful.
Not trying to argue any point (I mean honestly...), just trying to understand the German POV here, which is interesting.
So the German reader will read the English article (in English, not auto-translated), see the word billion, think “oh that looks familiar” and might assume it means the same as the German word Billion.
Machine translation makes quite a few mistakes, so I think if you have some decent knowledge of a language, you might be better off reading the original rather than a machine translation. At least my point of view from a couple of years ago. But it's also possible that machine translation has gotten WAY better in the last couple of years, I'm not sure.
I learned math of finance from a professor I'm pretty sure was quanting on the Street at the time.
When I told him one of the constants in our little trading game was off by a factor of 10x, he said prove it. My final talk was: "How to make 2.3 million dollars in 30 days using options."
The real lesson wasn't about pricing options. It was about being ruthless when the time came.
What I find amusing is that I'm categorically a much better functional programmer now than I was then, with professional F# experience with a lot of Clojure to boot, but they haven't responded back to any of my applications in the last three years.
I was a dropout for almost a decade, but managed to find work at big corporations somehow [1], and the entire time I was pretty obsessed with learning CS theory; partly because I liked it and partly because it made interviews easier. By the time I had decided to finish school and enroll in WGU, I already had roughly a bachelors' degree of education in CS anyway. And that's not just a pseudo-intellectual assertion on my end, I finished all my classes in a single term, most classes were completed in a week.
I don't think it's helped my job prospects much; I had already had senior level positions at Apple and Jet.com/Walmart Labs by the time I went back to school, and I suspect that they look better on my resume than a bachelors in CS from WGU.
(If you're curious, the reason I went back at all was so I would be qualified for a PhD program, which as far as I can tell universally requires at least a bachelors, and usually a masters, so I just wanted to check off a box. I'm not 100% sure why I wanted a PhD; I think my brain had fetishized professorships and CS research)
So it's tough to answer directly. I think for someone starting out today a degree from WGU is better than no degree, and I think the education they gave was "just fine". If you're ok with not working in Wall Street, it's enough to get you to the interview phase. If your goal is to work in Wall Street or to become a lawyer, it's probably better holding off until you can get into a "brand name" school.
I'm happy enough in my current position and they don't really care where I went to school, but I do believe that they had a firm requirement for a bachelors (I don't bother checking), and I like my current gig so it's probably worth it.
[1] I dropped out in 2012, which turned out to be just about the best time to drop out. It seems like everyone on earth suddenly had a smartphone and there was a shortage of engineers as a result.
I generally agree with the notion that talented folks are more likely to explore niche tech. Just be careful making the leap from "they prefer niche X" to "therefore they are talented".
Two anecdotes:
1. I'm an average programmer who likes niche tech.
2. My friend is 10x more talented than me, but he likes mainstream tech.
> Just be careful making the leap from "they prefer niche X" to "therefore they are talented".
I think I have a pretty good heuristic for finding curious and inquisitive people that doesn't rely on esoteric tech stacks.
My sense is that a lot of engineering teams have lost their way so the only mechanism they can use now are convoluted leetcode style interviews instead to filter out Senior Java Dev Candidate 43 versus Senior Java Dev Candidate 96.
I write a little about this here: Your Interview Process is Too Damn Long (and How to Fix It) (https://chrlschn.dev/blog/2023/10/your-interview-process-is-...)
When FB started to grow, it needed the HipHop VM[0] to make it scalable beyond PHP. I'd agree that PHP is a lowest common denominator and FB eventually needed to move beyond it to get to the next level.
Also, it goes Php -> HipHop (compile PHP to C++) -> HHVM (compile PHP to a virtual machine running ASM).
The domain probably isn't that complicated, but the infrastructure certainly was (and this was all pre-cloud so they built it themselves).
But yeah, FB's success came from a base level of keeping the site up and incredibly good tactics to drive growth and distribution.
Please do tell me then, what exactly motivates all the proposals for restricting trading periods or changing to auctions? All I have seen is that Jane Street and other make “too much money”. The reality is actually different they provide a service in 20 countries across all asset classes and do it for only $12B/year, that’s actually very cheap once the scope of their reach and operations are understood.
Please explain what is the motivation other than envy behind the numerous comments proposing some restrictions or claiming they aren’t worse or even better.
There are definitely academic sources that will backup my assertions however the irony is that my position somehow needs to be defended when it’s actually obvious to anyone who has spent a lifetime on the inside of the very system we are discussing and that somehow absurdist proposals about restricting market hours are accepted on face value without any rational or academic evidence whatsoever.
One can still choose to disobey the local government and be homeless. I'm not saying it's an easy choice, nor am I saying it's easy to be homeless. I'm just saying it's a choice.
EDIT: I'm not going to do this work for you and dig through all the places cited at the bottom to see if there's some source with that credible authority that proposes such a definition.
In contrast, if you're a prop shop that only accepts employee money, you're already dealing with your employees.
So there's no real reason for JS et al. to set large minimums, in contrast to non-prop hedge funds.
I think if I want to work at Jane Street now, I either need to finish my PhD, or get a Masters degree from a sexy school to "cleanse" my WGU degree (despite the fact that I don't think I know much less than the average Ivy League graduate, though obviously I'm a biased audience).
I'm not saying that I think this order of things is jolly good, but from the perspective of this kind of business sector it makes sense. Personally I avoid finance related sectors, a stint at a casino operator to get some experience on the CV was enough for me.
It still is a choice whether one chooses to work, as evidenced by those who chose not to.
Edit: the "luxuries" was slightly tongue in cheek and perhaps didn't land well. Anyway, you're free to move to the wilderness, hunt/grow your own food, build your own shelter, and do without the socio-economic system altogether.
Voluntary is not binary. If someone put a gun to your head and told you to eat shit you technically have a choice but it's not much of choice is it?
That would be a rather easy choice to make!
We aren't entitled to a roof over head and food etc, that's all I'm saying. We should be grateful when we happen to luck into circumstances (such as a job) that allow us to have roof/food. It's not something everyone has.
Saying we were involuntarily forced into having a job/roof/food feels very strange when compared with the people who weren't forced to have either of these three things.
Hecto = 100
Hectoliter = 100 litres
Hectogram = 100 grams
Kilo = 1000
Kiloliter = 1000 litres
Kilogram = 1000 grams
You’re not being empathetic by saying the opposite, you’re giving assholes with power an excuse to take those things away or make it harder to provide them to people.
I don’t understand why you’d say we’re being forced to have food and shelter, that makes no sense. We need food and shelter, which is why employment isn’t voluntary.
What does that mean even?
> it’s a tragedy that some people don’t have those things.
Agreed.
> I don’t understand why you’d say we’re being forced to have food and shelter, that makes no sense.
I was trying to take your position to the absurd extreme. You said employment was not voluntary, I took that to mean that you're forced to have food and shelter.
> We need food and shelter, which is why employment isn’t voluntary.
I don't know how else to explain this to you. As an objective fact, there are people who choose not to work. They do without. Some of them even manage to lead fulfilling lives. How can you pretend it's not a choice?
What a stupid, disingenuous fucking argument. You realize for the vast majority of people, food and shelter are top priority right? This is the foundation of Maslow's Hierarchy of Needs.
> What does that mean even?
Since I apparently need to explain the basics to you, entitled means "believing oneself to be inherently deserving". So yes, I think every human is entitled to food and shelter.
If socialism was a club, I'd take away your membership. Jesus christ.
Yes, and that is exactly why they choose to work! Because they want to have shelter and food! What don't you understand here!
> entitled means "believing oneself to be inherently deserving"
You strike me as a rather entitled person. What does "deserving" even mean?
I wish we could all have shelter and food without working and can't wait for you to solve this rather pressing issue.