Hacker News: suttontom

New comment by suttontom in "AI financial advice is surprisingly good, especially if you ask right questions"

suttontom — Sun, 02 Aug 2026 16:39:09 +0000

I still use YNAB 4 to this day on both the Android app and my desktop. Every couple of years the Dropbox sync stops working, but someone in the community fixes it and uploads an APK for others to download. It may just be that I'm used to it, but it still feels like the best combination of simplicity and design of any budgeting tool I've used.

New comment by suttontom in "The great wealth transfer reality check"

suttontom — Sat, 01 Aug 2026 04:00:53 +0000

> I’ve spent enough time on Reddit to know

Please tell me you see the irony in this.

New comment by suttontom in "A $500 RL fine-tune of a 9B open model beat frontier models on catalog review"

suttontom — Wed, 29 Jul 2026 17:52:27 +0000

It's kind of amazing how badly you've misrepresented Richard Sutton's position. This is the man who coined the term https://en.wikipedia.org/wiki/Bitter_lesson.

New comment by suttontom in "A $500 RL fine-tune of a 9B open model beat frontier models on catalog review"

suttontom — Wed, 29 Jul 2026 17:44:10 +0000

That is not how LLMs work. It isn't some intelligence that is separate from and also contains a database of knowledge that you can prune at will.

New comment by suttontom in "An Uncomplicated Man – A review of The Odyssey, directed by Christopher Nolan"

suttontom — Tue, 28 Jul 2026 18:39:42 +0000

This seems like a great insight at first but:

"What was the artist trying to achieve?"

How is that "disarmingly straightforward"? Does that mean it's a straightforward question for anyone to answer because it's obvious, or that it's straightforward for the viewer of the art because in that case the answer could be anything? If the former, that's an incredibly difficult question to answer and kind of nullifies the point of this entire exercise.

New comment by suttontom in "This July I Was Fired from Simple AI (A Deeply YC Company)"

suttontom — Sun, 26 Jul 2026 19:12:43 +0000

If you're having a manic episode or other type of mental breakdown, I hope you get help. If this is just how you typically act, and no one has told you yet, your way of communicating is bizarre and deeply off-putting. You will not be able to keep a job.

New comment by suttontom in "Demis Hassabis has a plan to harness AI safely"

suttontom — Tue, 14 Jul 2026 17:39:47 +0000

This is interesting, would being labelled "applied" be considered an insult to them?

New comment by suttontom in "Demis Hassabis has a plan to harness AI safely"

suttontom — Tue, 14 Jul 2026 17:38:47 +0000

I'm glad it's a few short years instead of a few long years.

New comment by suttontom in "Mark Zuckerberg's biggest legal nightmare yet could cost Meta $1.4T"

suttontom — Tue, 07 Jul 2026 19:04:12 +0000

I don't think the poster meant applying the death penalty to the corporation.

New comment by suttontom in "Programmers need to start meditating"

suttontom — Sun, 05 Jul 2026 22:10:10 +0000

Agree with you on exercise. But staring at a painting is to meditation what curls are to building your core. People all recommend the same type of meditation for a reason--so that you can close your eyes, focus on something, notice when your mind drifts, and bring the attention back. You didn't mention anything about awareness or noticing when you lose focus, so your advice seems to completely miss the point.

New comment by suttontom in "Claude Code is steganographically marking requests"

suttontom — Tue, 30 Jun 2026 22:49:27 +0000

It's crazy that you could actually use the excuse that since it's all vibe-coded, there's no way a human could have written it, so Anthropic bears no responsibility.

Meanwhile humans can pop in and leave little morsels like this and blame it on the model.

New comment by suttontom in "Better Images of AI"

suttontom — Mon, 29 Jun 2026 01:24:52 +0000

>These AI images also add to the public mistrust of AI, a growing problem for innovation in a field that is sometimes seen as biased, opaque and extractive.

Oh my, how would anyone ever have gotten that impression?!

New comment by suttontom in "I used Claude Code to get a second opinion on my MRI"

suttontom — Sun, 28 Jun 2026 18:10:49 +0000

Your instinct is correct, and in a lot of cases it's true. However, I've heard from enough doctors by now (a cardiologist, psychiatrist, and epidemiologist/former physician) that they use medical LLMs and find them extremely helpful, mostly as a way to either bring up knowledge they'd forgotten about or as a way to learn something new and then verify it. I'm extremely skeptical about LLMs in general and the connection to Gell-Mann Amnesia is apt, but I wouldn't necessarily write them off completely like that. There are experts using the models that find them genuinely helpful in their field.

New comment by suttontom in "Ford AI hiccups push carmaker to rehire ‘gray beard’ inspectors"

suttontom — Thu, 25 Jun 2026 17:06:54 +0000

How did it help?

New comment by suttontom in "The Doom Justifies the Valuation"

suttontom — Tue, 23 Jun 2026 00:37:37 +0000

It's a blog. He's using hyperbole. It's not a Supreme Court opinion.

New comment by suttontom in "Ask HN: What has been bothering you lately?"

suttontom — Wed, 17 Jun 2026 23:44:09 +0000

I'm asking genuinely, is there a connection between housing, education, and healthcare becoming so much more expensive and them also being the three parts of the economy that have the most government interference (in the US)? If so is it causal?

New comment by suttontom in "A backdoor in a LinkedIn job offer"

suttontom — Tue, 16 Jun 2026 07:13:51 +0000

I don't want to be cynical, but maybe spending hours every day using Claude has made some of us particularly attuned to picking this up. For some reason as soon as I read "The trap was in app/test/index.js," I instantly knew it was Claude. It's too bad, because there will obviously be some false positives, but it makes me immediately disregard the author.

New comment by suttontom in "Anthropic apologizes for invisible Claude Fable guardrails"

suttontom — Fri, 12 Jun 2026 18:40:23 +0000

You're wrong in lots of ways.

Some model cards do show regressions on benchmarks for newer models on specific tasks: https://storage.googleapis.com/deepmind-media/Model-Cards/Ge...

This wasn't a new model but updates to models backed by numbers being better can make the model worse: https://openai.com/index/sycophancy-in-gpt-4o/

The slight increases in performance/benchmarks may be just noise: https://arxiv.org/pdf/2602.07150

New comment by suttontom in "Workers are spending over 6 hours a week botsitting AI, fueling job frustration"

suttontom — Fri, 12 Jun 2026 02:00:09 +0000

I wouldn't agree with that. The issue with software is that the people you make things for are usually anonymous and you'll never meet them, but if you've ever built software that helped someone and you witnessed it, it feels really good.

New comment by suttontom in "LLMs are eroding my software engineering career and I don't know what to do"

suttontom — Sun, 07 Jun 2026 18:03:38 +0000

This is such a tired, meaningless argument. I've never seen a human in 10 years of professional software engineering at a large company ever so confidently, consistently create and send out seemingly well-reasoned code that's as wrong as what SOTA models using CC or Codex do. If a human did this, they would be fired or perpetually remain a junior who no one wants to work with.

Also, if a human does this, you can replace them and get a human who will not do it. The default for an LLM is to generate plausible-looking text that may or may not be completely incoherent. That is not the default for a human. Again, if you find that your colleague consistently fabricates APIs, you can hire someone who isn't crazy instead, but you cannot do the same with LLMs.