Programmer Moneyball (2016)

I wish more interviewers were like you! I happen to think there are a lot of developers and other tech employees out there who, for reasons you as the interviewer don't know, are plugging away at a company/role way below their capability or current skill level. Or, who are literal geniuses but don't happen to have MIT, Stanford or UC Berkeley on their resume. Capable of research and invention, but no PhD. People who just were never discovered, mentored and given the opportunity to get on a rocket ship. Raw talent sitting there writing CRUD apps for some dopey company just limping along, who sometimes don't even know how good they are.

Yet, when [the usual suspect] companies are hiring, they back a truck up to Stanford and just shovel new grads onto it. Or their automated filters (that desperately try to get the candidate flow from 6,000 to 6) eliminate anyone that doesn't have a narrowly defined "right background".

Better to just look for things like: grit, coachability/potential, focus, ability to reason about things across many orders of magnitude, ability keep a lot of aspects of a complex system visualized, and generally whether they can understand problems and come up with solutions. None of these involve playing "algorithm gotcha" or solving leetcode puzzles.

EDIT: Easier said than done, though. I don't have a practical solution to how to do this with a multi-thousand candidate flow for each job posting.

> We care about how fast you could learn new things that we haven't even thought of yet.

Are you accurately able to determine that in a conversation that lasts less than an hour? If so, I am really curious how.

I tutored people for years in college and I don't recall my first impressions being all that accurate. Accuracy would be even worse if I also had to guess the speed at which they would learn.

> If non-FAANG companies compete for the same prospective employees as FAANG employees, they will lose to vastly superior compensation packages. Fair enough.

Already this part is unclear. Baseball is different in another important respect: You have no choice about what you're doing. The Yankees, the As, anybody - is hiring you as, say, a pitcher. That's what you're good at, that's what you're doing.

But I chose to work where I do now, for far less money (the people hiring weren't sure but they suspected the previous salary I listed, which was true, was more than the head of their entire 100+ person department, several steps above me on the hierarchy) because of what I do.

What I do here makes a difference I care about, whereas say, working at an investment bank (as one of my ex-Google friends does) would not do anything except make rich people even richer.

If you're a big deal maybe you can carve a niche so that whoever it is that hires you, they're really hiring you to do whatever it is you do anyway. I can't imagine Bjarne Stroustrup feels what he does (at ms.com ie Morgan Stanley) is mostly about making rich people richer, Morgan Stanley presumably mostly lets him get on with C++ stuff. Likewise hiring Linus Torvalds wouldn't redirect him from working on Linux. If that was going to happen it would have happened a decade or more ago.

But for most of us in programming, where we work dictates what we do, and that influences whether I want a role I'm offered or not.

Companies largely succeed despite themselves when they find product market fit.

Moneyball wasn't about finding OBP - it's about finding what helps teams win (not making outs, ~2001) and simultaneously undervalues. The key thing here is this is dynamic - baseball no longer undervalues OBP.

Otherwise agree with what you say - some set math of Successful_Employee_Characteristics - What_FAANG_values, recalculated every couple of years, is how I would summarize the strategy.

> So the question, which traits are analogous to on-base-percentage? Take the FAANG-hire-resume, then subtract features (e.g. lose the top-10 university credential, remove a couple years of experience, ditch the years with a trendy package, etc). Which decreases the hiring cost the most with the smallest loss of long-term value to the hiring company? What data set would track/measure this and what could be deployed as a proxy?

The other question is: Why not simply match FAANG comp package by raising more money?

I'm being 100% serious here. Software is pretty much the only field which can scale almost infinitely and the only cost really is compensation (Compared to salaries, MacBooks are almost free!). Think about it, 3x comp to get a 10x engineer? Worth it.

Trust me, people are really religious about jobs.

To quote Homer Simpson: “It’s funny because it’s true.”

Whether the original commenter quoted by the author of this article (that quote should be proceeded by >> above, to make that clear) meant it in jest or not doesn’t change the fact that employers actually judge candidates with this exact criteria every day.

(Edit: clarity)

> ...need a lot more context for me to decide that the person who said it meant it for real.

Yes, of course. But the thing that ends up hurting a lot of people in a really deep way is the psychological hangs-ups caused when they conflate THEIR OWN self-worth to their employment status (not just employed or not, but also level and salary).

Any adult who has a solid self-image can tolerate hearing "you've failed as a human being" from derps all day long, what ends up causing damage is if they take it seriously, or much worse, tell it to themselves.

I've spent enough time in counseling over the past year to tell you that there are people out there who tie their identity so strongly to their jobs that this statement would definitely be true. There's at least one of them - that's me!

Part of why I'm in counseling is that I recognized this, and I want to change it.

Is it from people talking about themselves? If so, I'd default to believing it's honest. When people get a setback like that a huge amount of them believe they failed as human beings. Not doing so requires a large amount of introspection and wisdom.

If it's people talking about other people, without the emotional baggage, then I agree, it must be either irony or some evil one trying to control people.

Yeah, that's some "coffee is for closers!" phony business machismo.

I remember when Travelocity went through layoffs in 2009. A thousand people out the door, about a third of all employees. No me, and I had just received a critically negative annual review for the prior year.

It's extra funny, because the same movie has this wonderful scene I refer to often when I'm working as a consultant:

https://www.youtube.com/watch?v=fTjhHrcyiQI

TL;DW: "We're professionals, treat me like a professional. If it's not working out, say so, and I'll make it right or we'll shake hands and part ways on good terms". I used to provide clients with a "how to fire me" document because it made them feel comfortable hiring me in the first place. (Tell me, I return materials, pay me outstanding balances, here's the keys you change, here's how we wrap up the contract terms, etc etc)

I've been "fired" a bunch of times, and at times have had those same clients that "fired" me refer me to new clients. It's the same as with most technical work, I find: if you're not failing at least sometimes, you're not pushing the limits of what is possible.

Unless you hate web apps, then you are f**d.

Sigh. Here we are in the 21st century, implementing all our web page UIs in a TEXT EDITOR.

Have you ever tried to teach programming to beginners? 6 weeks is an absolutely bonkers timeline for zero-to-employable-as-six-figure-React-Dev (people were pissed at lambda school for dropping from 9 months to 6, and with good reason). I feel like step 0 of this would have to be “Be really familiar with computers to the point where you’re almost a programmer already”, because a lot of people take at least a couple weeks to fully grok things like:

- Where to write code

- How does a webpage actually work

- How webpages and other stuff get served

- How does the terminal work

- What does a compiler do

- What is a syntax error

- How scope works

- How functions and abstraction work (this one is often a huge mindfuck)

Then there’s the CSS and git stuff you mentioned. Beginners often find CSS super frustrating. The tutorials from different years recommend different things, and I occasionally find people trying to build pages using weird combos like position and grid or floats and flexbox. And then there’s git, which has a huge learning curve and is very difficult to pick up without someone to tell you “you’re trying to do this? Here’s what the command is called”.

In the end, even if you can figure all this stuff out that quickly (again, starting from zero) you’re going to be starting as a junior dev (You won’t be passing any tough technical interviews) and unless you’re in the bay area there’s no way you’ll be making close to six figures. And finally:

> you don’t need to be good at programming

If I smell this attitude on someone, I’m not hiring them. If someone is new, applying as a junior, and is willing to learn, that’s fine. But when I hire someone junior, I know I’m going to have to do a lot of work to train them up and fix their stuff as they get acclimated to working in software. And if they don’t follow directions, learn from their mistakes, or show a desire to improve because they’ve internalized the mindset that “you don’t need to be good at programming”, I’m going to be quite disappointed in them.

I went from 2XXk to xxK when I moved from managing a recruiting office to software development then to 1xxK within a year of software development.

I didnt follow this exact path but while "have a skill set people will pay for and know how to negotiate" is a drastic over simplification it is possible and realistic.

You also didn't mention where to work. In my country a React job will get you 50k a year at most as a junior, and almost nobody recruits you if you don't have a master's degree.

This sounds like a recipe for 'devops' people. They don't know a single thing about anything. It's like their brains are full of copy and paste bullshit.

They're almost as useless as "CyberSecurity" graduates.

Generalizing here, of course.

Yes, I think it probably pays to looks at the top school in each region for this reason! I knew a lot of very smart people at University of Florida because it was essentially free with the bright futures scholarship.

It does tend to be a shorthand. Obviously, there is no reason that someone on a fixed-length contract would be intrinsically different than others. In the corporate line-of-business world, contractor often means staff augmentation and most staff augmentation firms are selling by volume, not quality.

We hire FTE software engineers to be autonomous creative problem solvers and leaders.

We hire contractors to be a pool of labor for clearly defined and menial tasks that would be a waste of FTE time.

Interesting that all four of these posts (including this one) are from 2016. Have things been changing?

He's a very good writer, and he's good at finding interesting things to discuss.

Would you buy a book of collected blog posts? Would you be willing to pay, say, 99 cents for a nicely printed and bound copy of a given blog post in pamphlet form if it also came with a handful of "shareable" (read: disposable) ones?

I used the title from the url stub. I didn't realize it was a different title than the first heading on the page. No disrespect intended.

So true!

But is this just another example of “the market can stay irrational longer than you can stay solvent”, meaning, is there a severe mis-allocation of capital into building large, expensive, inefficient software engineering organizations? If so, why is the market not “correcting” for this?