thinkr42

This entirely. You can't eat using "good feelings" / "impact" alone.

Genuine curiosity here and trying to temper my own selective memory : does it seem like skepticism is growing on the AI front in the news recently?

This mostly makes me wonder about the varying definitions of NLP. Most business folks I’ve worked with in this field don’t even know the question to ask or prompt- the implementation details of models like GPT are the easy part.

I've personally encountered this quite a bit circling around tech, in most of the cases I've seen it was due to _deeply seated_ insecurity more than anything else. Usually it corresponds with folks who find themselves in a position that they're a poor fit for but for whatever reason can't move on. They bounced into a position they can't exit and their 'value add' is applying this sort of tactic under the guise of being helpful externally.

Though verbose and somewhat strange at times, one thing I love about SQL is that the query statements read like a set definition from set theory. That declarative nature is pretty powerful IMO, sure there are hiccups but it is a different way of thinking.

Skeptical to say the least, only because I think this person would likely get a Nobel Prize if this was proven to the be the case. Bell's theorem is huge in QM, usually introduced very early in the Quantum track. It'd be awesome if there was a flaw, but the implications go far beyond quantum computing then.

Company "culture" always makes me a little skeptical, particularly when they're used for hiring decisions. I can't help but think it is just another way to discriminate against folks - I mean if you're saying some one isn't a culture fit what are you really saying about them?

I get it if you're telegraphing that you just don't think they'll fit with you team, but having seen a lot of culture interviews in practice, they seem to be more socially acceptable code for "they're not gender conforming in profession, preference or practice", "they're not white", or what have you.

"For the love of God, don't use Feynman lectures to learn physics. That's something you read after you know physics, for relaxation and conceptual stuff..."

Agree with this 100%. To learn math or physics you have to do it, experiment, and all that.

So some of these graphs remind me of benford’s law, something you see in a lot of number sets. https://en.wikipedia.org/wiki/Benford's_law

What’s so powerful for me is that my old company’s “brand” is exactly this style

I’ve been in a workplace where we supposedly followed this. It was often used as a veil for passive aggressive toxicity. There was a lot wrong with that place, and this certainly isn’t the cause of it, but I’ve seen this used as a weapon to do some really shady stuff in the workplace.

That is something I hadn’t thought of before

It is remarkable how frequently “unreasonable effectiveness” is thrown around.

I hadn't thought of this to be honest, but I do faintly remember the IT / Dev team using the "not technically feasible" line a bunch whenever Marketing would come through with stuff. Used to get me angry quite a bit and I saw it translate a bunch to lost revenue and increased costs.

I don't know how much of this got sent out of Marsh and up to Sun though.

I saw this a bunch in my next job after Marsh, which was analytics consulting. I'm not sure about Marsh doing this (at least internally), but who knows with Sun.

I worked at Marsh from 2005-2015 in both corporate and the store. We were always skeptical of Sun, the CEOs that would cycle through didn't grasp the basics of grocery, and as an organization we failed to invest in basic infrastructure that would reduce our overall costs which are huge in the 2% margin grocery business.

At the store we had registers with an OS from the 80's, a common thing if you look around, but we never invested in developing anything beyond the 80's unlike our competitors. We didn't ever implement a JIT style logistics program, which with massive in store staff cuts led to poor merchandise orders which in a perishable business kills you. Management in general was _highly_ skeptical of new technology and prone to spending a bunch of labor hours on repetitive, easily automated tasks (by my accounting at the time, I could have automated more than a million dollars of labor costs with a week of effort - something I brought up with management and demonstrated but was ultimately shot down, I wasn't some super genius - we had folks literally copying and pasting for 8 hours a day).

Marsh was a good company with good people, several of those in the article I knew from my time in corporate. We had folks that would kill themselves working 60 hour weeks to keep a store afloat. It sucks to see what had happened to them.

Do you think that some of the hiring practices are out of whack as well? I've heard it claimed that sometimes the government treats hiring as welfare employment programs - which adds a whole new spin on this.

If only we could use it to travel around the world instantaneously.

Used this to detect tax fraud for a state government and uncover a breach (by poor suppression techniques) of medicaid data. This little law is extraordinarily useful!

This is a relatively small problem - that sure is one ugly red knob on the side. I don't think it jives with the design at all and looks cartoonish. Otherwise, seems like a decent upgrade

Is this sort of analysis technically allowed by Twitter's ToS and Privacy Policy? Don't get me wrong, what you post on social media is public and that's that, but I know there are sometimes interesting rules around this sort of analysis.

This is great, but think "Probably is in there" should be replaced with "Might be in there".

Surely there's a Chart.ly (or alternatively 314Chart.io) that is in the works somewhere - disrupting this market using tensor flow deep learning networks to produce the best pie charts possible.

Absolutely agree. This weighting is pointless considering the interdependent nature of the items being weighted.

Completely agree. The examples are incredibly narrow and likely document and context dependent overall.

This is awesome!

Byzantine hiring processes in the government itself doesn't help this problem. This isn't a politica statement, if you compare USAJobs to an average tech company you'll see an over emphasis on certifications and the like. There are reasons for it, but it doesn't help when decision makers or project managers don't know what real success looks like or how to even measure it.

You are absolutely right about that!

I respectfully disagree. While I understand where you're coming from, I don't agree with your distinction between an analyst and a scientist. Given the data scientist's typical compensation and expected experience, there should be a higher bar set for them that does include developing solutions from base. I understand the use of utilities, but far too frequently I find people who rely on packages to do their work don't really understand what they're working on (they often don't realize the underlying assumptions that the package writers made for them either). With your description of the tasks for a data scientist, I would label this as a Data Analyst's work if I was hiring one.

I could of course be wrong and have a bit too narrow of a view from my particular subfield.

Completely agree. There's a lot of nuance in these algorithms, they're not as cut and dry as simply calling a package method and oftentimes they aren't optimized to your use case. I work in Machine Learning, specifically on NLP, and it is really obvious when interviewing potential employees who knows what SVD means and who just know the NumPy function. Most "data scientists" I've interviewed fall in the latter category.

edit-This is of course completely anecdotal experience.