This is what infuriates me the most about healthcare in the US. I give someone my card, they enter the numbers with no indication of whether or not everything will be covered by insurance (they assume you just know).
Then, a month later, I receive a bill indicating a routine procedure that I assumed out of ignorance was covered by insurance (since they took my card and entered the numbers without saying anything to the contrary) for $500 (or, god forbid, more).
If I knew a salad at a restaurant were $200, I probably wouldn’t order it. There is no basic transparency in medical billing, and that needs to change.
It’s actually even worse. If you (or the Dr’s) could reliably tell how it was even going to be coded (aka categorized/identified) in the system in advance, it would already be a huge step up.
Then you’d only have a handful of different prices you might have to pay.
Specifically for A/B or A/B/N testing, you can use a beta-bernoulli bandits, which give you confidence about which experience is best and will converge to an optimal experience faster than your standard hypothesis test. Challenges are that you have to frequently recompute which experience is best and thus, dynamically reallocate your traffic. They also only works on a single metric, so if your overall evaluation criterion isn’t just something like “clickthrough rate”, this type of testing becomes more difficult (if anyone else knows how multiple competing metrics are optimized with bandits, feel free to chime in).
Beta-Bernoulli multi-armed bandits (BB-MAB) are definitely a good way to get started on a Bayesian version of A/B testing where you've the additional benefit that your population is dynamically allocated to the most performant option (actually this dynamic allocation makes it similar to interim analysis [1] rather than vanilla A/B testing).
There are some caveats though - and I mention these from the experience of running such solutions on a large scale in production. First, BB-MAB can't adapt to context by design. They only look at click/no-click behavior across the population. So, if your population has two distinct segments - youth and elderly - who behave very differently wrt purchases, the BB-MAB won't pick a different winning advt. per group; its blind to these groups.
The solution is to use something like a contextual MAB - which assimilates user features (or whatever you might throw at it) into the MAB. There are simple ways to adapt simple MABs to the contextual setup [2] (in my experience, these can also be effective) but, of course, the literature in this area is wide and deep.
A second caveat is that if the ratio of the size of the pool of advts. to the number of impressions is high, the BB-MAB won't converge or converge to a good optima; the search space is simply too large relative to the data. In cases like this it becomes important to begin with the right Beta priors, instead of the standard recipe of starting with a Beta that looks like a uniform distribution.
Can you point to a book or other resources that support this claim? Interested in learning more. Always looking for another reason to hate cars as a New Yorker
I'm reading "The Geography of Nowhere" right now, it covers much of this in a reasonably entertaining fashion, although its not chocka-block with references.
It does make some predictions that haven't dated well though (like hitting peak oil in 2023)
As someone who lives in New York and has seen the episode you’re referring to, it is not embellished at all. New York might be the hardest city to find a public restroom that I have ever been in. Sometimes buying something isn’t even sufficient grounds for using a bathroom - a business might only have an “Employees Only” bathroom in the back.
This is so funny, I was just doing this with ChatGPT to apply for jobs. If you make me fill out a bullshit form where I fake being passionate about your company, I’m going to get a chatbot to write it for me. End of story. I just need a job, why do you care if I’m passionate about building CRUD apps for you?
I am a pretty hard to read person. I don't excited about switching jobs even if its for more money. I generally have a monotone voice when interviewing.
Some time ago, after an interview I aced, the recruiter called me and asked "why I don't seem to be interested in the position". I was like what? Apparently, because I didn't sound excited about working for an insurance company they were on the fence. The manager I interviewed with at the end was obnoxious, saying things straight from a script. I don't think I can work with people who act like "excited puppies". I'm a damn adult just trying to program and learn new stuff, but mostly put money in my bank.
Maybe I would be excited about working for some space program......but even then there is always the day to day.
I'm not convinced of that, there's lots of people who are diligent but quiet about it. Passion works great in some job contexts, in others it can lead to exploitation, cult-like behavior, groupthink that results in a terrible product etc.
Obviously you want staff that like their work enough to be fully engaged with it, but fake passion is bullshit that's often foisted on us by marketing/HR drones. Think of how people who work at Subway stores are called 'sandwich artists' when the reality is that they'd probably be fired if they deviate even slightly from the approved recipe.
I don't want to work with people who are primarily driven by passion tbh, because they're likely to either burn out or be intransigent when there's a difference of opinion. I just want people to be friendly and not robots.
In my current role, I have to sometimes write T-SQL or PL/SQL stored procedures in 80's/90's tech, sometimes scripts in Groovy, sometimes starightforward Java and sometimes code that interacts with relatively more modern tech like Kafka. Do I prefer some over others? Yes. Do I spend equal amount of care towards quality of code? Also yes.
A good programmer will always try to do a good job irrespective of their level of passion with the stuff they are working on. Does enjoying a specific piece of work more produce better code? Maybe because I would have more fun writing it. But I doubt most of us gets to do the exciting work everyday. Ironically, if you are relying that much on passion, you might get those who cut corners on unexciting work. Rely instead on good programming skills coupled with professionalism.
A professional gets shit done. With my higher experience, I have a better understanding of what provides business value and don't work on it more than that, because it is not valuable.
I am passionated about technology, programming and system design. I have about zero passion for writing documentation and good pull requests - but I do it anyway, and I like to think I do it well.
If I only did the parts of my job I had passion for, I wouldn't be a very good employee. If people only worked for companies they were passioned about most companies would have so much more trouble hiring (who is passionate about working for Wells Fargo? Doing SCADA work for a flour factory?).
We at company XXX are building a solution that does a,b,c.
We want senior level programmer with a preference given to a candidates having these specific skills / domain knowledge.
The job is in the office, (no) need to travel. Requires (or not) clearance / degree / license / etc.
We pay this much (range is ok) and offering such and such benefits.
Please explain why you are a good candidate for this job.
Skip enthusiastic, passionate, hard working, woke, tolerant, etc. etc. bullshit because everyone is asking the same crappy questions here and gets the same crappy answers and it is all meaningless.
I have also been out of work for ~8mo for various reasons, and while I’m sorry for your situation, it’s heartening to hear I’m not the only one. I’ve only actually gotten one tech screen in which I was asked a LC “hard” question, which I didn’t pass (I’m pretty good with “easy” and “medium”, but come on, hard?).
So far, I have one other interview from an internal referral, but haven’t heard anything back from the ~10-15 other applications I’ve sent out. I’m getting a bit discouraged and feel similarly as you - it would be nice to take more time off (maybe I don’t actually love working in tech?), but I have the same worried as you. Feels good to vent in this thread, though :)
I felt a bit bad about posting here, because the thread is ostensibly about layoffs, while I quit voluntarily, so any feelings I'm feeling about the situation now are my own fault :)
But it's somehow comforting to see I'm not the only one marveling at the sudden slowdown. It eases the inevitable fear that it's just something unattractive about me or something I'm personally doing wrong.
I didn’t listen to this, but the essay is full of straw men and contrived arguments about how utilitarianism is a bad moral philosophy. While utilitarianism is not perfect, this essay just regurgitates others’ criticisms of it and provides no useful alternative. As an aside that bothered me (but not related to the content directly), the tone of the article also wants to be scholarly so badly, which made it really long-winded and hard to read.
>> Indeed, this very essay, which I’d long planned to write some version of, is coming out now because the same effective altruist organization is offering a $20,000 prize to whomever gives the best critique of effective altruism this month.
I think this sentence tells you everything you need to know about this guy’s philosophy. The article also ends with him suggesting, as an “alternative”, to give a money to fundamentally utilitarian/effectively altruistic sources (AI safety research) and to fund projects that he’s interested in under the premise that these underfunded scientific fields are “awesome and epic”.
What you call strawmen are just example of where the philosophy falls apart. i mean, how else do you test a philosophy other than throwing all possibly scenarios at it and see what comes out?
Those two things aren’t mutually exclusive at all. It‘s very possible for the algorithm to get content to be amazing while still posing massive national security concerns.
I'm not talking about the algorithm quality. I'm talking about considering something so threatening "extremely entertaining". It has to be some domain (moral maybe) where those concepts are mutually exclusive.
I’m sure not everyone came here to hear tech book recommendations, but I will add another vote for Designing Data-Intensive Applications by Martin Kleppman. It’s one of the best tech reference books I’ve ever read. It manages to explain SO much while requiring so little prior knowledge from readers.
Another book that is relatively new that I loved was Designing Machine Learning Systems by Chip Huyen. I worked in productionizing ML systems for 3 years and this book equips you with exactly what you need to do so. It does a great job of explaining the whole ML modeling pipeline and some of the commonly overlooked nuances that can cause your models to fail spectacularly in production. I will be referencing this book for years to come.
I'd suggest searching HN for other reviews of the book since it comes up pretty frequently in these threads, but personally I liked that it covered a lot of ground quickly and fairly deeply, and the content remained engaging for me throughout. Some specific things I liked about the book were its coverage of SQL isolation levels and how they work, different ways to coordinate separate systems when you want atomicity (like 2pc and queue-based synchronization), failure modes to keep in mind when dealing with distributed systems (like unbounded network delay and split-brain situations), and references to a whole bunch of production-ready software that solve distributed computing challenges in various ways depending on the trade-offs you're looking for.
Overall I felt better equipped to make intelligent decisions about systems design after reading it.
Awesome list, thank you for sharing! Curious, though - how do you structure your weeks so that you can consume 50 books a year? That would take me a while!
For me, it's with a Kindle. I bring mine everywhere as it's so small (Kindle Basic) I can put in the back pocket of my pants. Whenever I have to wait for something, or I'm taking a break, I just open it and the book is where I left it. 5, 10, or 50 pages, they all add up. Most books are around 600-1000 pages on it.
I can finish a book this way in about 3-4 days. But my reading quota has lowered this year as I was reading wuxia (chinese) books where each is about 27k to 30k Kindle pages.
I read in fits and spurts. I read a bit after work, sometimes at lunch, and definitely before bed each night. Weekend, if I'm not otherwise busy, I'll read for a couple hours as well.
Then, a month later, I receive a bill indicating a routine procedure that I assumed out of ignorance was covered by insurance (since they took my card and entered the numbers without saying anything to the contrary) for $500 (or, god forbid, more).
If I knew a salad at a restaurant were $200, I probably wouldn’t order it. There is no basic transparency in medical billing, and that needs to change.