I'm in an online degree program in mathematics in my forties and this temptation is very real. The LLMs have memorized every textbook and every exercise so it's easy to have the kinds of conversations that before I could only have with TAs during office hours, and skip the mental struggle.
At least in my most recent class, it's also wrecked the class discussion forums that I previously found very helpful. By the end half the students were just slop-posting entire conceptual explanations and exercises, complete with different terminology, notation, and methods than the class text. So you just skip those and look for the few students you know are actually trying.
Someone mentioned this kind of thing in the npm breeches last week. You get desensitized to non-standard “official” urls and it makes it easier for phishers.
> Pipelining, cache misses, branch prediction, multiple cores, even virtual memory are all completely invisible to C programs.
Are there languages where these are visible? Any time I've seen them discussed it's been in terms of writing C/C++/Java that understand how they work, e.g., the disruptor pattern padding variables to a 64-byte cache line.
> C teaches you an abstraction of computers based on the PDP-11.
Hmm, I'd wonder how a C designed today would look based on modern architectures.
Live trading systems are in C++ or Java, but like others have said Python is used for day-to-day research, sometimes R if the particular quant likes it.
I never saw any Fortran in my time at Bloomberg in the 2010s but allegedly there was still a lot of it running in the nether regions.
My advice here won't be one-size-fits all and there's far too much to cover against your company's mission to just give you a hot list of tools I like.
The biggest line of battle is going to be Google Workspace vs Office 365. I'm fine with them both, but not fine with giving them both to everyone. The problem is, if the majority of your company's documents and drive data is fractured over both MS and Google's clouds, it's going to be a nightmare to manage. Choose one and make them your identity provider and life partner. I understand the finance team probably needs a stand-alone Excel license, though.
The other very important part of standardization is hardware. This is something I had to get a bit ruthless about, because managing a fleet over an entirely remote trans-border workforce is expensive if you expect quick turnaround times on onboarding and an immediate replacement for that VIP who just cracked their screen.
I ended up setting up a rapid logistics operation using drop shipping.
Sales, marketing, etc get a 13" Macbook Air M1.
Developers get a 16" M1 Pro because they justifiably need the performance.
No, we don't do custom orders. Not big enough to have spares of all kinds in stock (Apple supply chain lag was a real drag last year). Apple's MDM systems make those machines useless bricks to anyone outside the client's company. (If this sounds like something you need, reach out to me!)
I guess I'll just add that my biggest pet peeve with SaaS vendors is how they want you to pay for the super-duper-expensive package just to get SSO / SAML. That's ridiculous. It's not a cost center for you and it reduces both our risk. If I don't see at least "Sign in with Google" in your cheapest plan, I'm already pissed off as a customer and not going to be your change agent.
At least in my most recent class, it's also wrecked the class discussion forums that I previously found very helpful. By the end half the students were just slop-posting entire conceptual explanations and exercises, complete with different terminology, notation, and methods than the class text. So you just skip those and look for the few students you know are actually trying.