[walkabout]

I would have to think a really long time to come up with an answer to “describe a hard problem you’ve solved” at all. I might have to consult notes. After that, to make it an interview-friendly narrative, I’d have to fill in a lot of details with plausible speculation because god knows I don’t actually remember (and I don’t really trust memory very much, anyway)

The joys of age plus extremely-poor autobiographical memory.

It doesn’t help that what I think of as actually hard, day to day, is shit like documentation from FAANG companies lying to me and wasting a whole fucking day of my time, terrible error messages, mismanagement of upstream projects, bureaucracy making a two-day task take four weeks, and, in some common ecosystems, awful design of core tools and major libraries, or horribly wasteful library churn. “Hard problems” are nothing next to bullshit problems.

[anal_reactor]

My career isn't long, about five years, but so far I haven't encountered a technical problem that wouldn't be trivial. Like, "just follow whatever ChatGPT says" trivial. The real challenge are people problems - corporate work teaches you that the real tough shit is surviving being surrounded by idiots.

[walkabout]

Not a lot comes to mind that really falls between “basically trivial” and “you’re gonna need to hire a team of PhDs” or “you have accidentally asked me to write a custom distributed file system, and you probably don’t want to pay for that”

Which is to say that usually the hard stuff is an accidental request that nobody actually wants to pay for, so the only thing “hard” about it is recognizing that the client/stakeholder has wandered into territory they really ought not.