It might be shocking but there is like a whole world outside of computers and even outside of your job.

I'm fully aware, I'm trying to develop some skills here and there in various different fields that interest me

Thing is I don't know on which one to focus on as a backup, I wanted to hear some ideas from the HN community

It is hard to know what works well, as everyone is different. I would look at things that seems fun. Fixing stuff, from bikes to engines? Physical therapy? Teaching kids? And so on.

For some you may be even able to leverage your software skills.

Here is an idea of something that will be the last to be replaced by AI: Public Speaking.

If you have time, start developing your public speaking skills. This will give you the edge over other engineers within your company, especially if you know how to use it. Additionally, you'd be able to survive if you are layed off by either moving to an adjacent field (pre-sales, sales, CTO...etc) or just start attending and presenting at conferences. Hope this helps

You mean "avatar-free 3D speaking" where you are confirmed to be real by your presence in a room?

What if I want a whole world inside of computers?

Get professional help

I can't afford it, I'm unemployed, AI took my job

As a counterpoint, I have enjoyed that last several months greatly. I have been able to solve real, long standing, issues quickly and efficiently.

Not sure what to recommend for you though.

> I have been able to solve

Is it you though?

Last week "I" have chased down a really sneaky performance bug in a shutdown routine of one of our applications, if it went to prod it would cost us a lot of grief. How did I do it? Well, I didn't, I gave Claude access to kubectl, described the behavior and how to trigger it, and went for a coffee. After 10 minutes of didly-dalling I took a sip of coffe, then copy-pasted 100% spot-on response to a JIRA ticket and sent it to developers for a fix.

I did maybe 2% of work there.

The business had problems that needed solved. I talked with folks and gathered data to be sure what the problem was and what a solution looked like. I then pointed an agent at the proper part of the system with a goal. It gathered data, came up with a plan, created a MR with my name. I submitted the MR, got it approved, made sure it worked in production.

I didn't 100% think through every decision in the implementation but I am 100% responsible for ensuring the problem was solved. What tool I used to solve the problem doesn't diminish my work.

Same, I had a very weird issue with tailscale VPN for a very long time, no amount of updates or googling fixed it

I created here something similar to how intranets work, where I can easily define a service as public or internal in k8s through the ingress class

If it's internal, it's accessible only through tailscale, if it's public, it's accessible by anyone, now for some reason internal services have insane delays or very very bad bandwidth, never understood why and couldn't figure out how to even debug this

Ran codex, gave access to the whole node, ah yeah just lower the MTU to this value, lo and behold it fixed everything

The other day I got a cheap thermal printer with the only way to use it is through an app, I thought ok I'll try to reverse engineer it, try to understand the protocol, then write a script to print, few days later and progress was very slow, I thought let's try to let an LLM so it

It reversed the entire protocol, wrote me a web app where I can print whatever I want, I got what I needed, but feel zero accomplishment

> I got what I needed, but feel zero accomplishment

Just a personal view, but I feel great when I get that result. I had an effing problem that was a pain but is now fixed after a short conversation. I get to check and confirm that results are good for me. And it gets done without spending hours with random tech support ("did you try reinstalling your OS"?). My 2c.

> then copy-pasted 100% spot-on response to a JIRA ticket and sent it to developers for a fix.

I get these and the fix ai suggests is often wrong or hallucinated!

Yeah, that's annoying. I've cross-checked with another engineer before filing the ticket. I agree that just slop-shotting someone is just a bad mannner.

The best idea I've come up with is quitting my job, living on my savings and indulging in my hobbies until I run out of money and then hanging myself in the woods.

I've been thinking about going into biochem. Seems like an interesting problem space where, in the past year we've seen massive progress in our understanding of molecular biology.

> My question is, what now?

The answer you're looking for is not on Hacker News.