Hacker News: nrub

New comment by nrub in "Muse Code and Muse Spark 1.2"

nrub — Wed, 05 Aug 2026 23:25:04 +0000

Right, but the point is that you can't conclude that a model is necessarily bad because it's not hitting the same scores on benchmarks. I just don't agree with lacker's conclusion, because their logic doesn't seem to consider that. Scoring lower on a benchmark doesn't strictly mean they have a bad model, but it may be the case. Like you said it's one data point, but being the easiest, and obviously most gamed, means you should probably weigh them less heavily.

New comment by nrub in "Muse Code and Muse Spark 1.2"

nrub — Wed, 05 Aug 2026 21:28:16 +0000

Or they spent time optimizing their model to real world problems they're facing and didn't waste time trying to game a benchmark.

New comment by nrub in "Hands-On with the AMD Ryzen AI Halo"

nrub — Fri, 10 Jul 2026 22:45:40 +0000

That's true, but that's just one model that has some easier setup in the ecosystem. It's not easy to get a consumer gpu that has 128Gb of vram, so while the Halo chip isn't as performant, the ceiling for running larger models is higher.

New comment by nrub in "Write code like a human will maintain it"

nrub — Fri, 10 Jul 2026 22:42:37 +0000

In opencode you can get granular, allow `git log *`, deny `git add *` and `git commit *`

New comment by nrub in "Show HN: Microsoft releases Flint, a visualization language for AI agents"

nrub — Wed, 08 Jul 2026 19:13:20 +0000

> simple chart specs can be reliable, but generated charts are often of low quality due to reliance on system defaults; - complex chart specs with explicit details can produce good-looking charts, but they are verbose and agents can struggle with reliability

N of only a few of us working on an analytics agent, I don't think we've been finding this to be the case. We've been impressed with just how good LLMs (even smaller open weight models) are at using Python and R for visualization. Often any shortcomings go away if we iterate a bit to about ambiguity. Are there any threads of research that could better support this claim or highlight where issues might be?

New comment by nrub in "Microsoft Office 2019 and 2021 for Mac view-only conversion"

nrub — Sun, 31 May 2026 03:37:57 +0000

Just because they don't know how to use a computer well, doesn't mean you can't teach them.

New comment by nrub in "CISA Admin Leaked AWS GovCloud Keys on GitHub"

nrub — Tue, 19 May 2026 15:33:38 +0000

I no longer keep my dotenv files in plaintext. I use `sops` to keep an encrypted env around and you can use tools like direnv to make them available to your shell while you're working. Obviously the LLM could print any of these secrets, but it's less likely. Additionally I find that at least claude seems to avoid reading the dotenv. And lastly, don't make any local secrets that important. Limited scope, dev accounts, etc.

New comment by nrub in "If AI writes your code, why use Python?"

nrub — Tue, 12 May 2026 22:30:38 +0000

No, they're not there. Did you read my comment? It's not just flawed or limited implementation, those are things that python just doesn't have.

New comment by nrub in "If AI writes your code, why use Python?"

nrub — Tue, 12 May 2026 17:31:49 +0000

Yeah, that's cool, but it would be almost completely unnecessary if python just had actual private methods/classes/properties. It's a lot like pydantic, which is completely unnecessary if you had strong typing.

New comment by nrub in "If AI writes your code, why use Python?"

nrub — Tue, 12 May 2026 17:26:54 +0000

Yes, but: a) they're a second class citizen, not guaranteed to be used in whatever niche of the python ecosystem you find yourself in and there's already an n+1 problem with multiple type checker written by third parties, rather than having 1st class language support tool that's consistent. You're not going to get it by default, you're usually going to have to do some configuration (and maybe bike shedding) to get it working; b) they completely negate the idea of python being "easy to read", your code is now littered with `if TYPE_CHECKING:`, `Literal`, `TypeAliasType` and any number of workarounds needed to make your hints work out. Unfortunately the syntax was just not designed with typing in mind, and I think it shows; c) the idea of "hinting" rather than enforced type checking means you have no guarantees that a type is what you need it to be, you have to do a lot of boundary work to make sure the edges of your code are coercing things to the right type. While I love pydantic and find it to be an excellent library, to me it's the kind of code smell you get in languages without strong typing. Also you're going to get a lot of spurious type errors along this path as well;

I will gladly use python's type hints, it's a whole lot better than nothing (IMHO better than typescript), but in it's current form it will always fall short of a language that was designed with strong typing in mind.

New comment by nrub in "If AI writes your code, why use Python?"

nrub — Tue, 12 May 2026 17:26:43 +0000

You joking?

- strong typing - real concurrency (heaven forbid you want a background task without having to spool up an external message queue and worker) - immutability - limitations in error handling (sort of just typing really) - limitations in nullability (also typing) - memory layout is usually hidden or abstracted away - no actual private methods or classes

That's far from a complete list, but maybe you're taking for granted the typical pythonic conventions that many practice. It requires a ton of work to design and architect python systems of any non-trivial size for maintainability and understanding. No language is perfect, but there are plenty of languages that make supporting complex systems easier than python.

New comment by nrub in "Bring Your Agent to Teams"

nrub — Thu, 23 Apr 2026 14:52:59 +0000

You have literal Stockholm syndrome. Not joking.

New comment by nrub in "Results (Don't) Speak for Themselves: A Case for Documentation"

nrub — Tue, 10 Sep 2024 16:33:24 +0000

RCA with FM radio, https://en.wikipedia.org/wiki/Edwin_Howard_Armstrong

New comment by nrub in "Devin: AI Software Engineer"

nrub — Tue, 12 Mar 2024 18:28:11 +0000

I've seen a few suspect benchmarks for recent announcements of LLM releases. I'm sure they made an attempt at an honest benchmark, but until there's an independent assessment and benchmark (preferably multiple) you have to assume that there's bias in anyone's self published benchmarks like this.

I'm guessing it's a fine-tuning of some existing LLM model or API, but this largely seems to be an "agent" and UI that includes some SWE like workflow coding to allow more complex requests to be asked than just an LLM could provide.

New comment by nrub in "Where is Noether's principle in machine learning?"

nrub — Fri, 01 Mar 2024 16:19:12 +0000

There are promising methods developing for Physic's informed neural networks. Mathematical models can be integrated into the architecture of neural networks such that the parameters of the designed mathematical models can be learned. Examples include learning the frequency of a swinging pendulum from video, amongst more advanced ideas.

https://en.wikipedia.org/wiki/Physics-informed_neural_networ... https://www.youtube.com/watch?v=JoFW2uSd3Uo

New comment by nrub in "Going beyond ‘do you know of any open positions?’"

nrub — Wed, 12 Jul 2023 18:27:24 +0000

Do what you want with some of those tips, maybe not necessarily good advice. But if you're searching, try these search queries replacing `%s` with your desired role(s):

``` search_strings = [ 'site:boards.greenhouse.io %s', 'site:jobs.smartrecruiters.com %s', 'site:jobs.lever.co %s', 'site:.wd5.myworkdayjobs.com %s', 'site:icims.com inurl:jobs %s', 'site:careers.peopleclick.com %s', 'site:apply.workable.com %s', 'site:jobs.workable.com %s', 'site:apply.workable.com %s', 'site:.taleo.net %s', ] ```

Most of those ATS systems are super annoying. I actually give preference to employers using more modern systems that have up to date user experiences, but regardless, it doesn't hurt to scrape and search all the big ATS players.

New comment by nrub in "Executable Examples in Go"

nrub — Sat, 10 Jun 2023 15:51:02 +0000

Yup

New comment by nrub in "Plotting the source code “TODO” history of the most popular open source projects"

nrub — Mon, 17 May 2021 18:55:12 +0000

One option if you're on github, https://github.com/marketplace/actions/todo-to-issue

New comment by nrub in "Evolution Is the New Deep Learning"

nrub — Thu, 22 Mar 2018 23:33:49 +0000

The paper linked above does directly address the case of multiple experiments occurring in the same context. They address this with hill-climbing over those 180 different variations. The use of a bayesian linear regression takes place of the exploration found with Thompson sampling.

New comment by nrub in "Prodigy: A new tool for radically efficient machine teaching"

nrub — Fri, 04 Aug 2017 20:24:28 +0000

They have some hard comparisons in some of their earlier blog posts on how spacy compares to the other popular open source NLP libraries. In my experience it has been much easier to use and faster than things like Stanford's library or NLTK. In general it's aimed at production or commercial use, whereas the other libraries I typically hear mentioned are aimed at a more academic audience.