Hacker News: justinlivi

New comment by justinlivi in "Falsehoods Programmers Believe About LANs"

justinlivi — Sat, 05 Sep 2026 23:20:40 +0000

I feel like the better of this is maybe "will" instead of "can"

New comment by justinlivi in "The people who actually want AI to replace humanity"

justinlivi — Sun, 31 May 2026 15:45:44 +0000

Any supposed "AI BDFL" will be controlled by a human. The base concept is inherently flawed.

New comment by justinlivi in "Using AI to write better code more slowly"

justinlivi — Tue, 26 May 2026 00:47:38 +0000

I find myself spending on average more time in LLM review/resolution loops than it would take for me to write the code by hand. Partially because once I'm in the flow I write very very quickly and the code pours out sometimes faster than I can write. But also because the LLM code on the first few tries is generally really really bad. What I find interesting though is that spending the time to personally review and direct the LLM through several iterations of review and revision on average results in higher quality code written in about the same time as I would have written it. This might be particular to me, but seeing several interations of someone else's code helps me better understand holistically my objective as opposed to whatever happens to come out of my flow-state consciousness.

New comment by justinlivi in "Embarrassingly simple self-distillation improves code generation"

justinlivi — Sat, 04 Apr 2026 21:16:06 +0000

From my own empirical research, the generalized models acting as specialists outperform both the tiny models acting as specialists and the generalist models acting as generalists. It seems that if peak performance is what you're after, then having a broad model act as several specialized models is the most impactful.

New comment by justinlivi in "Claude Code daily benchmarks for degradation tracking"

justinlivi — Thu, 29 Jan 2026 23:41:05 +0000

Compacting equals death of the session in my process. I do everything I can to avoid hitting it. If I accidentally fly too close to the sun and compact I tend to revert and start fresh. As soon as it compacts it's basically useless

New comment by justinlivi in "The Timmy Trap"

justinlivi — Fri, 15 Aug 2025 17:49:12 +0000

I think AI skeptics have a strong bias to assume that human intelligence fundamentally functions differently from LLMs. They may be correct, but we don't have a strong enough understanding of human cognition to make the claim in as uncertain terms as the skeptical argument is unusually made. The training methods between human learning and machine learning are obviously fundamentally vastly different as are the infrastructure-level mechanics. These elements are likely never going to align, though with time the machine infrastructure may start to increasingly resemble human bio hardware. I bring this up because these known vast differences may account for a significant portion of the differences in expected output from human and machine processing. We don't understand the fundamental conceptual "black box" portions of either form of processing well enough to state definitely what is similar or dissimilar about those hazy areas. Somewhere within that not-well-understood area is what we collectively have vaguely defined "intelligence." But also within that area are all the other aspects that both humans and now machines are quite good at - prediction, fluency, translation. The challenge of lexicon and definition is potentially as difficult a task as is sharpening the focus of our understanding of the hazy black-box portion of both machine processing as well as human processing. Until all those are better defined I don't think we have a good measure for answering the question of machine intelligence either way.

New comment by justinlivi in "Show HN: LogoFox – fast logo maker"

justinlivi — Thu, 23 Nov 2017 15:30:56 +0000

The problem with this model is that if you aren't able to do due diligence yourself due to technical restrictions, how are your end users supposed to overcome the even greater technical restrictions on due diligence that using your service imposes?

New comment by justinlivi in "Moral Machine"

justinlivi — Tue, 04 Oct 2016 03:37:15 +0000

I took the same sort of dispassionate approach, valuing the lives of the passengers above all else and staying the course otherwise. I was disappointed to discover the parsing of the results had no room for such methodology. Based my entirely algorithmic approach it was determined that I favored youth and fitness.

New comment by justinlivi in "Unraveling Möbius strips of edge-case data"

justinlivi — Sun, 12 Jun 2016 20:51:10 +0000

I'm curious to see how HIPAA regulations interact with this type of research. I would imagine it would be seriously limiting (though my actual knowledge of the laws is severely limited, so maybe not?)

New comment by justinlivi in "Introduction to Metaprogramming in Nim"

justinlivi — Tue, 07 Jun 2016 01:48:27 +0000

As someone who's never coded in either go or nim, this seems to me to be the exact opposite ideology of golang. The metaprogramming is very cool, but I imagine sharing a code base that utilizes it heavily is a nightmare.