nathan_douglas

    Location: Ohio, US (ET)
    Remote: Yes
    Willing to relocate: No
    Technologies: Rust, JS/TS, Python, Terraform, Ansible, Kubernetes, Flux and Argo, a whole bunch of other DevOps/Infra stuff, random backend stuff, Linux sysadmin, whatever's on the plate.
    Résumé/CV: https://ndouglas.github.io/resume/resume.pdf
    Email: See résumé.
    GitHub: https://github.com/ndouglas
    Website: https://darkdell.net/

I tend to be the guy that gets locked in a room with a shitty job no one else wants to figure out, and then I come out two (days|weeks|months|years) later with grayer hair and wild, bloodshot eyes, but with the mission accomplished.

I like science and I like weird shit. I'm trying to grow and get as good as I can be, and I'd love to work on something interesting and really difficult with really smart people.

I'm an absolute idiot but I have some good points too. Strong interests in complexity science, ideonomy, and a few other quasi-heterodox things.

Same. I'm also a big fan of the wild poster [1], too.

[1] https://eblong.com/infocom/maps/zork-1-poster.jpg (big)

As a side note for anyone who sees this, I received an unsolicited email from "Harold Anderson" (harold@signalvapor.org) but signed "Chris", trying to get me to buy an AI service (nspire.ai).

nspire.ai 's company page [1] lists as their head of sales "Chris Kinnard". Perhaps this is coincidental. Perhaps nspire.ai, and their head of sales, has nothing better to do than to spam people. Doesn't say much for the quality of their product.

Anyway, I wanted to share this scummy, shitty, spammy behavior, just in case anyone now or in the future happens to Google™ Theo, nspire.ai, AI career coaches, etc.

[1] https://nspire.ai/company/

    Location: Ohio
    Remote: Yes
    Willing to relocate: No (but see below)
    Technologies: Kubernetes / Terraform / Ansible / Serverless / AWS / Rust / Bash / PHP / Python / JS / TS / C / Clojure / etc, happy to learn more
    Résumé/CV: https://ndouglas.github.io/resume/resume.pdf
    Email: See résumé
    Site: https://darkdell.net/ 

I'm a generalist software engineer with most of my experience (14YOE professional) in backend, platform, infra, and DevOps stuff. I'm currently employed and love where I am and what I do, but I intend to relocate to the EU in 2028 and my job (federal contracting) is incompatible with that goal.

I'm very interested in deep learning, formal verification, and compliance automation, but also artificial life, complexity science, robotics, cybernetics (as in Beer), computational social sciences, network science, transit, city planning, etc.

I use Claude Code happily at work, as directed, and seen significant (though not unqualified) productivity gains. I've used it to varying degrees on different personal projects. I'm avoiding it now on personal projects (to retain and build skill). I can work anywhere along that spectrum.

I enjoy learning about what other people are working on, so even if you're not hiring or not interested in hiring _me_, I'd love to connect. I've had some genuinely transformative conversations with people I've met on HN, and I'd love to have more.

*EDIT*: I mean "I'd love to connect" IF you're not just trying to sell me something - especially your shitty AI résumé- or interview-prep services, you hacks.

Good idea!

On another note, I dig your box of cards. I've recently re-encountered the Commonplace book [1] idea in multiple places (and when two events pertaining to the same object of inquiry occur simultaneously, I always pay strict attention), and this is kind of a similar idea.

I've had wikis before but they always became more about the engine / storage / infrastructure / etc, and I like your lightweight vision.

[1] https://en.wikipedia.org/wiki/Commonplace_book

Are you finding the French decks helpful? I'm also trying to learn French (not using spaced repetition _per se_ but Pimsleur [which does use spaced repetition, really], InnerFrench, and reading [currently reading _Le Trône de fer_]).

FSRS is really cool. I'm trying to use it and a modified flashcard system to learn more abstract computer science and higher math. I hadn't considered it as a way of learning Chess - that's really interesting. I'm thinking about expanding my system to cover ear-training, birdsong recognition, a few other things like that.

I ended up giving up on it because my right shoulder is damaged from rheumatoid arthritis. Go where I could not, anon291. May your reeds be clean and your bellows free of mildew.

Really dig the sound of the first one, though all of these are really cool.

You're probably good if:

    - you're actually listing a job
    - the job pays money
    - the job pays a sane amount of money for the region in which you're hiring
    - you're actually trying to hire someone for the job
    - you don't demand they jump through stupid hoops like write an essay about Stendhal's views of modernity or your favorite unrolled loop from the GNU standard library
    - you don't come back every month for the next three years advertising the same job without hiring anyone for it
    - the business model isn't obviously delusional ("we're going to feed the global poor with AI, we just need to, uh, figure out some niggling little details")
    - you reply to inquiries and applications

Every day we stray further from the light of our lord and savior, Dr. Nicole Forsgren...

If you run it in a cluster, does it become a Lemon Party?

There are no incentive structures (besides possibly "posterity") to encourage anyone to see past their noses. In fact, hardly anyone at any level of any organization, public or private, is able to operate with a real longterm, sustainable outlook. They'd get shitcanned for trying to plan ahead, even if they were intellectually equipped for that.

That's actually blowing my mind. Thank you so much for making this comment.

@#$% that's impressive. A little above my budget. I appreciate your response. I read some of your other comments and will follow your career with interest.

lol, yeah, that's an excellent example.

I wish I had a stronger background in behavioral economics/psychology or some good connections in academia. The "instantly stop playing" thing seems really extreme and thus _really_ interesting. I'd expect people to think "nah, screw you buddy" but not leave entirely.

Unfortunately, while I follow a number of people on LinkedIn who work in the computational social sciences area, I'm not aware of any of them that actually follow me. I'd love to get their opinions on it but most of them are I think more in the area of economics than psychology or social psychology.

What's your GPU setup like?

I'm doing a vaguely similar thing - I have a 10" rack minilab [1] and I've vibe-coded an MCP server that runs in the cluster to introspect, etc, but the main longterm goal is to set up some ML pipelines and maybe work toward formal verification via TLA+ or smth. (_not_ vibecoding that... I'm thinking of moving into AI formal verification or compliance automation as a career move.)

I have a separate amd64 server with an RTX 2070 Super - which is obviously old and low-powered. Useful for some general ML stuff, but I don't think it's sufficient to run any non-trivial modern LLM.

I'm thinking about upgrading that GPU, but haven't committed to it or even really thought that hard about it.

[1] https://clog.goldentooth.net/

Ah, I think there was a German word for that idea, right? Lebensraum?

My guess would be that they're in a profit vs. loss mindset and thus less inclined to be charitable, community-minded, etc. You're basically forcing them to reason purely selfishly and then wondering why they don't donate.

For instance, I wonder if (all other things aside) Nicky Case would have the same problem with Parable of the Polygons ( https://ncase.me/polygons/ ).

I think this is distinct from the cognitive vs. emotional empathy issue you raise.

Oh man, that's super cool. I wonder if techniques like this will be used to investigate other games, possibly contributing to a sort of "family tree" of games kind of like the ones we have for linguistics with conjectured languages like proto-indo-european, etc.

Superpowers + worktrees works really well for me. Superpowers works well at building comprehensive plans for implementing large features or refactors, and asks the questions up front, and then worktrees provide a safe place to actually perform that work.

It's not perfect; I've had some issues with Claude Code forgetting where it did things ("oh... it's not working because I'm not in the right directory"). I think it needs some architectural tweaks to function more reliably.

Wow. I did not know this. I'll bring it up in my organization.

> Despite the substantial amount of blood on the walls and throughout the house, investigators found no biological evidence on Henning, Birch, or in the stolen car. To explain why, prosecutors relied on the testimony of Dr. Henry Lee, who at the time was the director of the Connecticut State Police Forensic Laboratory. Lee testified that although Carr had been brutally stabbed numerous times, including a cut to his jugular vein, the blood from his wounds had spattered in an "uninterrupted" fashion, meaning there was nothing between Carr's body and the wall. Asked whether the assailants would have had blood on their persons, Lee said, "My opinion is maybe."

> As part of his testimony, Lee used a crime-scene photograph of an upstairs bathroom that showed two towels by the sink. Lee testified that the towels had tested positive for the presence of blood.

...

> The towel and Lee's testimony about it took center stage. In 2008, the towel had been tested and no blood was found. In addition, there was no report indicating any previous test. Attorneys for Henning and Birch argued that Lee had testified falsely and that prosecutors had compounded that false testimony by using his misstatement in two separate closing arguments.

https://exonerationregistry.org/cases/12854

Cool! I ran this against a personal project and this found a good number of issues. Looking into each of them, they seem to be valid. The run was fast, and the output looks clean and actionable. Thank you!

Oh, I love it! I really loved the art style on the water in this game.

I don't know anything about anything TBH. I'm just an interested dude. My day job is just working with microservices and stuff, nothing related to AI or ALife.

Probably the closest thing I have right now is a p5js sketch with a genetic algorithm here: https://bitterbridge.github.io/p5js-sketches/robbie-the-robo... (takes a few minutes to start working effectively and probably has a ton of bugs).

I've started making some other sketches like https://bitterbridge.github.io/p5js-sketches/monsters/ to set up this kind of roguelike idea for demoing some more of these things, like Goal-Oriented Action Planning, but I haven't made much progress yet.

I also have this Lisp rules engine VM I created over the winter break, https://github.com/ndouglas/longtable/ , which might become the core of a more complex reasoning system. IDK though.

So little time, so much to do :)

Thank you so much for the recommendation! I think that's right up my alley and I'm gonna grab it right now.

Disclaimer: I'm not an expert in anything, let alone linguistics or language acquisition. Just reporting what I've found.

Effective fluency (C2+) seems to be really, really difficult without immersion.

Anything past A1 seems to be really difficult without dedication and active effort. Pimsleur is my foundation and I think it's debatable whether that will get you out of A1 by itself.

I do Pimsleur (practical listening and speaking and reading), InnerFrench (listening comprehension, roughly A2 where I'm at right now but I think it progresses through B1 and B2 as he speaks faster and rephrases things less for ease of understanding), and reading (_A Game of Thrones_, currently, which is handy because I've read the books in English before and can use that as leverage).

I also watch some French films, but that's less helpful at present. I can't always find good French subtitles for my (pirated) films and because I watch a decent number of foreign films (especially Polish, shoutout to Wojciech Has enjoyers), the trying-to-learn-French part of my brain mostly shuts off. This is just a place where I need to put in more effort.

I have a lot of other French language podcasts and some YouTube channels cued up, but I'm not ready for most of them.

And this doesn't really address two of the major components, which are writing and speaking French. I don't have much going on there. I plan to add 2-3x/week iTalki next year so that I get feedback from fluent French speakers.

So TL;DR: your skepticism is absolutely justified and it's an uphill battle, a significant challenge and a sink of time, money, and willpower. I've been doing it for a year, on top of three years in high school and another year in college, and I think it's getting harder, not easier.

So we'll see.

I'm trying to get to B2/C1 in French and intend to move to France in 2028. Over the years I've picked up a little Spanish here, took a few years of German there, etc.

Recently I read _Erec and Enide_ [1] and it was really cool to be able to find the original Old French version of it and read large parts of it (not the whole thing) and find it so much easier to read than Early Middle English like the _Ancrene Wisse_ [2], etc.

One of the things I've really appreciated about LLMs is to be able to ask about the divergence of the Romance languages, e.g. "why does 'y' mean 'there' in French and 'and' in Spanish?" and get a legible response. It's really enhanced the learning experience by taking seemingly arbitrary differences and situating them in historical contexts, etc. I think it makes more connections somehow and helps me build fluency faster.

IDK what my point is, I just find this stuff fun to think about, even if you're not a French language learner. I'm gonna have to dig deeper into this site, thanks for sharing.

[1] https://en.wikipedia.org/wiki/Erec_and_Enide [2] https://en.wikipedia.org/wiki/Ancrene_Wisse

This is cool. I envy you for being able to work on projects like this.

There's a spectrum of intellectual ability, and I tend to find myself interested in things far lower on the scale - insects, worms, amoebas, etc. I have a strong interest in GOFAI - pathfinding, goal-oriented action planning, expert systems, etc, and have played a lot over the past decade or two with implementing these things in roguelikes - simple 2D-ish world, sensory stimuli, etc.

So the recent stories about teaching a wad of human neurons to play _DOOM_ and this specific model are pretty interesting to me, even if, again, it's a few orders of magnitude in complexity higher than my focal point.

Are you aware of any similar work that's going on at lower levels with simulations like this? Did TRIBE v1 follow from simpler models or was it "greenfield", so to speak?

(Apologies... I figure I could Google my questions but sometimes I learn more from an HN comment than I would in hours of trying to parse a graph of scientific papers and news articles.)