I'm guessing I'm over twice your age but my situation is similar to yours in a couple aspects. What I decided to do is focus on exploring the new capabilities that AI will allow individuals to do (especially those of us who can code). There are a lot of ppl who are justifiably frustrated with the current state of the tools. But don't get discouraged by that. I've seen A LOT of tech waves. It's still early in the AI era. I won't suggest any particular approach but recommend giving it some thought.
BTW, being an old timer, I also hated the leetcode thing when it became a widespread thing after Google adopted the practice in the wake of The Joel (Spolsky) Test. But you know, years later when I had to go through some interviews, I did spend a few months studying it and it turned out to be pretty fun. But passing technical screenings didn't help me land a job... only connecting via my network seemed to make any difference.
After Coursera/Udacity/EdX discontinued courses that I wanted to take, or removed access to ones I only partially completed, I switched to buying classes on Udemy. I completed only a handful of many purchases, and the quality level was okay-to-mediocre but better than nothing, so I got more value out of Udemy than Coursera.
I also found that Youtube videos are just as informative as Udemy classes, but they're not always as well structured.
The MOOCs had some pretty cool/interesting university classes that don't exist anywhere else. It's a shame those videos weren't preserved where we can access/purchase them without attending the college.
I agree, some of the worst employees I've seen were hired that way.
I haven't hired anyone recently but btwn 10-20 years ago I did hire a lot. Of course we reached out via our network of connections but that gets tapped out fast, so you have to rely on job postings. It was always hundreds of applicants per opening. Back then it wasn't 1000's but it might as well have been because I didn't have enough time to sift through them all. That's ok, you can just approach it like "the dowry problem" (also known as the secretary problem [1]).
But the job market and hiring is way worse now, and it's pretty horrible for job seekers atm.
As mentioned elsewhere here, Ocaml and the ML family are often cited as easy to implement other languages, due to pattern matching, enum variants, sum types, etc. Since newer languages like Rust and Swift copied those features, you might be more interested in them since they're more popular than older ones like Ocaml and Haskell. Personally I like the syntax sugar in Swift.
> In the absence of Western tools like ChatGPT and Claude, many Chinese universities have begun deploying local versions of DeepSeek on campus servers to support students. Many top universities have deployed their own locally hosted versions of Deepseek. *These campus-specific AI systems–often referred to as the “full-blood version” of Deepseek—offer longer context windows, unlimited dialogue rounds and broader functionality than public-facing free versions.*
This makes so much sense. Are there any U.S. universities doing this?
A caipirinha with vodka is called a caipiroska. I name my computers after caipirinha varieties. Once at the office my laptop screensaver was showing "caipiroska" and a Russian guy exclaimed, "what's that?!?" Apparently "piroska" has a certain connotation in Russian that's nsfw.
More context at:
"Moonshot's Kimi K2 uses a 1T-parameter MoE architecture with 32B active parameters and outperforms models like GPT-4.1 and DeepSeek-V3 on key benchmarks"
<https://www.techmeme.com/250712/p11#a250712p11>
BTW, being an old timer, I also hated the leetcode thing when it became a widespread thing after Google adopted the practice in the wake of The Joel (Spolsky) Test. But you know, years later when I had to go through some interviews, I did spend a few months studying it and it turned out to be pretty fun. But passing technical screenings didn't help me land a job... only connecting via my network seemed to make any difference.
reply