Hacker News: aselimov3

New comment by aselimov3 in "My Mental Model for "Is It Worth Automating?" Has Changed"

aselimov3 — Thu, 21 May 2026 13:09:17 +0000

Good reasonable take. I think the tradeoff on AI automation is trading skill development, brain muscle growth, and technical understanding for speed. You have to balance it so you are net faster but also growing. Otherwise you are just indistinguishable from any other prompter.

I’m unconvinced by people specializing in “agentic engineering techniques” like managing agent swarms/spec writing/harness tuning. Regardless of how complex your set up is I don’t think you are actually growing your skillset/problem solving abilities if your only interaction with code is prompting. I also think those “skills” have a finite lifetime as each generation of model and harness requires different ways to hold them.

Eventually if the AI promise is meant to happen (which I’m skeptical), LLMs will be able to fully manage their own swarms/write their own specs without requiring our input. Only spending time learning how to get better results from today’s AI is basically a waste for your long term growth.

New comment by aselimov3 in "Show HN: I built a private, manual 0% balance transfer tracker"

aselimov3 — Thu, 21 May 2026 11:19:25 +0000

Missing the link

New comment by aselimov3 in "Stop paying $360/year to access your own email history"

aselimov3 — Wed, 20 May 2026 00:56:41 +0000

Site broken for me

New comment by aselimov3 in "Google Antigravity Built an OS from a single prompt"

aselimov3 — Tue, 19 May 2026 21:10:32 +0000

lol paid 1k to play doom with worse performance probably

New comment by aselimov3 in "Coding on Paper"

aselimov3 — Tue, 19 May 2026 20:51:03 +0000

Man that’s a hard one because personal browsing for things like banking would require a GUI. But I could use a separate machine for that. This is a pretty decent experiment to run.

New comment by aselimov3 in "Going full AI engineer, not touching code anymore"

aselimov3 — Tue, 19 May 2026 16:03:10 +0000

I was mainly just speaking to the point of the article with regards to ability to think about code/solve technical problems. I think most EMs I speak to would say their technical skill set is degraded from years not writing code. I think if you just review code without writing it, your technical skills degrade regardless of whether you are reviewing junior or LLM code. In both cases you maybe gain architecture design and soft skills but you are definitely not improving your technical skill set

New comment by aselimov3 in "Going full AI engineer, not touching code anymore"

aselimov3 — Tue, 19 May 2026 15:01:29 +0000

> Sometimes I get the "you're right!" response, but often it will also explain why it made the decision it did, and it's rational enough that I accept the new approach.

I think this is kind of what I’m worried about. Referencing Karpathy, an LLM can basically convince any one of anything. Doing this enough times and your opinions just become the LLM opinion. Same with problem solving approaches. I see this for myself so I reserve time to formulate my own solutions. Maybe my solutions are worse but at least I’m training the muscle that may lead to me outperforming the LLM in specific spaces

New comment by aselimov3 in "Going full AI engineer, not touching code anymore"

aselimov3 — Tue, 19 May 2026 14:56:33 +0000

But the specs can’t be fully detailed or else they would just be code which is the point the article above makes. You can certainly limit where the LLMs have freedom by writing more detail but obviously that defeats the purpose of LLMs. What I mean more broadly though is more along the lines of baked in solutions. The LLM will generally prefer certain architectures/libraries/patterns. If you develop a spec alongside an LLM (which I think most people do) you will get pushed towards those. You kind of close yourself off to other ways of solving a problem.

You can stop this by detailing exactly what solution you want but I think continuous leaning on the LLM will lead you into a confined space.

New comment by aselimov3 in "Going full AI engineer, not touching code anymore"

aselimov3 — Tue, 19 May 2026 14:50:28 +0000

That’s a good point but I think this article above would push back on this point slightly. I certainly am able to do a lot more with LLMs because it can produce a passable solution in a lot of places but I’m not sure it expands my solution space really. I tend to separate what I’m able to build and what I’m able to solve. I can build a frontend with an LLM but I don’t think I’m able to solve frontend engineering problems.

New comment by aselimov3 in "Going full AI engineer, not touching code anymore"

aselimov3 — Tue, 19 May 2026 14:34:58 +0000

I think the main issue with this approach is that your solution space eventually narrows. You go from formulating unique solutions to flagging things that look wrong and then just picking an alternative LLM generated one. I’ve noticed this recently as all my side project work is being done by hand. If I do ask an LLM another question about something small it will offer solutions but doesn’t offer the solution I think makes sense in the architectures I’ve written. Slowly your brain just gets trained to mid thinking like an LLM. At least that’s my theory and I think there was some study that showed that interacting with an LLM at the start of a problem significantly narrowed the solution space to something sub optimal

New comment by aselimov3 in "Type Out the Code"

aselimov3 — Tue, 19 May 2026 14:24:22 +0000

This is very true. I haven’t tried the deleting and retyping from memory but I plan to start. I’ve gone back to setting aside blocks of time where I use no llms and read documentation to solve my problems like a caveman. Even if some of the issues are for relatively small things that aren’t generalizable I feel like it still stops my brain from rotting and improves my general reasoning and problem solving capabilities

New comment by aselimov3 in "Coding on Paper"

aselimov3 — Tue, 19 May 2026 12:44:29 +0000

It feels like with this monitor the only thing you can really do is a terminal. Would really help with focusing I think. Less colors/distractions. Although I’m sure hn would work fine on it haha

New comment by aselimov3 in "Coding on Paper"

aselimov3 — Tue, 19 May 2026 03:19:00 +0000

If only I could justify the 2k. I feel like life quality would improve significantly if I had this. Do you use another monitor for calls for work?

New comment by aselimov3 in "I'm banning AI from my life for all human-to-human communication"

aselimov3 — Mon, 18 May 2026 18:13:29 +0000

I did the same thing! https://www.alexselimov.com/posts/ai_confession/

I don’t think it’s even necessarily safe to use it as a grammar checker because it will offer phrase changes that are hard to deny. Then you slowly lose your voice as a writer.

New comment by aselimov3 in "The Borrowed Brain Problem"

aselimov3 — Mon, 18 May 2026 15:54:17 +0000

What on earth is this?

Handling the great code forge fragmentation

aselimov3 — Mon, 18 May 2026 13:10:49 +0000

Article URL: https://www.alexselimov.com/posts/forge_fragmentation/

Comments URL: https://news.ycombinator.com/item?id=48179325

Points: 3

# Comments: 0

Show HN: Track 10x bathroom tile developer status across code forges with Hugo

aselimov3 — Mon, 18 May 2026 00:20:55 +0000

Article URL: https://github.com/aselimov/-hugo-unified-git-activity

Comments URL: https://news.ycombinator.com/item?id=48174378

Points: 1

# Comments: 0

New comment by aselimov3 in "How to Start Investing with $100"

aselimov3 — Sun, 17 May 2026 20:38:03 +0000

Actually a really nice post and helping me to pull the trigger. I’ve been waiting to invest because I feel like a crash is coming but I should stop feeling that way

New comment by aselimov3 in "'No way to prevent this,' says only package manager where this regularly happens"

aselimov3 — Sat, 16 May 2026 11:27:28 +0000

Yea I’m a big fan of rust but it does feel uncomfortable to see my dependency blowing up to the hundreds when I build

New comment by aselimov3 in "'No way to prevent this,' says only package manager where this regularly happens"

aselimov3 — Sat, 16 May 2026 01:39:19 +0000

What are the actual guarantees that go/Rust make that Python/npm don’t? It seems like it might just be that Python/npm are juicier targets? I’m starting to try and avoid all third party packages