I’m not a retard, but I’m nowhere near people like Sphere either. I’m working hard on getting better and making up the time I wasted when I was a kid, but I get the feeling that I’m running a race in which I don’t even know if the finish line will still exist when I get there.

I will try to find places where I’ve written about this in more length, but I think the key thing to understand is that this is not a good comparison. I also wasted a LOT of my youth and was never especially truly gifted like someone like Sphere is. It took me way more time to get good than it did for him. That’s completely OK, because my competition wasn’t Sphere, it was people applying for the jobs I was applying for.

If you can program at all, and you have some time to study the fundamentals of computers and how they work + get a good mental model of programming as a practice, you will be fine. You will catch up to them, even surpass them within your own niches (sphere doesn’t know everything! he knows nothing about ML for example!)

So I guess my point is that it shouldn’t be discouraging to see these folks – there’s many teenagers around the world who could probably outcode me in their sleep, but I have other things. Some of these things I didn’t have fully realized at 22, but there were some seeds already. It allowed me to do fine over time!

I’m trying to get entry level positions/ internships, but it feels impossible right now and with AI I expect it to only get harder in the future.

Yeah, the job market sucks hard for entry level and internships. OTOH, the hardest part is getting your foot through the door. There are a few things you can do, including networking, or looking at entry level jobs you might want to apply for and checking their requirements to reverse-engineer what you have to learn (and then build a few toy projects around that to maximize your chances). But I agree that it is tough out there.

Is getting “cracked” even a realistic goal when there are 14 year old that run laps around me? I know you sometimes talk about the brilliant kids you talk with on here and whenever I see that I wonder how I can compete with people like that

I touched on this above, but certainly you can get cracked much later on in life. My best friend had never programmed until 27 and was basically stuck doing phone support at a bank, and 4yrs later he works on self-driven cars (writing Java services for data collection, ok, but still!). So yes, it is possible. But also, cracked is a complicated and sometimes silly concept. You can grind data structures and algorithms all day, or do competitive programming but that doesn’t really make you cracked. I think my definition of cracked is someone who becomes familiar with computers, can program them to do the things they need them to do, is a great systematic thinker and can reason from first principles, someone who is curious and tries to learn because it pays off in the long term. And that’s something you can become at any age, and certainly can at 22

If you do think that proceeding makes sense, how should I direct my energy? - Should I get a masters degree or just try to get a good job/ build something?

If you can get a masters without starving or being in trouble, I think it’s safe to get one. Just don’t cruise, and study on top of what they give you, and develop your interests. Take side quests on the way, etc… But a masters who consolidate your skills most likely. You can always build some smaller projects on the side and if something comes up, drop out on the way

  • What should I pivot into?
  • What should I do to stand out to get a job?
  • I’m very interested in ML (I know cliche), but it feels like the space is extremely saturated and top heavy making it even harder to compete in.

This is hard to answer, because it depends on what you’re maximizing for. Is it getting paid, doing something you enjoy, etc etc, and the tradeoffs you’re willing to make. One thing’s for sure is that it can change, and it’s ok to go after the money first and suck it up and pivot later. It all depends, but as far as a specific field, I think it’s good to ask yourself: are you conscentious enough to grind something you don’t actually love? if not, I would take a bet on yourself and do the things you actually have interests for. If that’s in the AI/ML, that’s totally fine. There are tons of opportunities for applied ML, ML automation stuff etc… The loop of “learn, build, learn” can be applied there too and side projects will help. Same with the tricks of looking at job requirements for these positions and reverse-engineering what you need to learn and/or be familiar with