Hacker News: andyfilms1

New comment by andyfilms1 in "Vacuum-Form Signage"

andyfilms1 — Thu, 11 Jun 2026 13:42:54 +0000

Gridfinity is fun, but just about any other organizational method is easier and faster.

New comment by andyfilms1 in "LLMs are eroding my software engineering career and I don't know what to do"

andyfilms1 — Sun, 07 Jun 2026 16:53:36 +0000

Additionally, without the knowledge of how you got from A to B, you don't know what else is possible (or impossible.) In the process of doing something manually, you may stumble across a particular setting or effect that creates something you never even considered. And now, that is knowledge you can use on the next project.

Show HN: One-Shot Program Generation Through Direct Memory Diffusion

andyfilms1 — Fri, 05 Jun 2026 14:55:09 +0000

This is a POC of an idea I've had for a while. Rather than using LLMs to generate code, what if we could just write a software image directly into memory?

At the end of the day, programming languages are an abstraction to make development easier for humans. Using an LLM to write code is like having someone else chew your food and spit it into your mouth--it gets the job done, but it would be faster to just cut out the middleman.

To that end, I'm using a diffusion model to generate images for a Von Neumann VM. The model treats the VM machine state as an image (literally saveable as a .bmp) with pixels representing bits. To reduce nondeterminism and noise, several direct x0 predictions are made and their logits are averaged, with the result thresholded into binary pixels.

Because the diffusion model may still make occasional pixel-level errors, the image stores important logical bits multiple times, and the decoded value is chosen by majority vote.

It's not perfect (and right now it's just capable of basic arithmetic) but the result is an executable image, generated in one-shot, stored in a small 72x72px .bmp.

I'd love to hear some thoughts on this!

Comments URL: https://news.ycombinator.com/item?id=48413376

Points: 1

# Comments: 0

New comment by andyfilms1 in "Launch HN: Rudus (YC P26) – AI for concrete contractors"

andyfilms1 — Tue, 02 Jun 2026 21:59:02 +0000

I was going to say, I don't understand what this does that Revit doesn't already do better. I guess it's a fun demo, but I doubt this was a problem that needed solving.

New comment by andyfilms1 in "Werner Herzog in conversation with Paul Cronin (2014)"

andyfilms1 — Sat, 30 May 2026 21:03:50 +0000

Absolutely. One of the major purposes of art is as a way to communicate things that can not be put into words. And if that's the case, how could one ever hope to create it with a prompt?

New comment by andyfilms1 in "Tech CEOs are apparently suffering from AI psychosis"

andyfilms1 — Wed, 27 May 2026 18:10:23 +0000

Correct, it's addiction.

New comment by andyfilms1 in "Claude Code as a Daily Driver: Claude.md, Skills, Subagents, Plugins, and MCPs"

andyfilms1 — Wed, 27 May 2026 14:24:09 +0000

The comment wasn't about CC specifically. If you rely (like, can't ship without it) on any model that you don't control, it's not really your product. If Dario decides to increase pricing 500% because it's Tuesday, and you can't work without CC, you really have no choice but to open your wallet.

New comment by andyfilms1 in "Microsoft starts canceling Claude Code licenses"

andyfilms1 — Fri, 22 May 2026 18:56:45 +0000

I expect in the next year or so, we'll stop seeing headlines like "Anthropic buys $15b of compute from SpaceX" and we'll start seeing headlines like "Uber's AI department licenses GPT 6.2 as the foundation for their internal model," or something like that.

Smaller companies will have departments that distill larger models into something more specifically manageable and useful for them. At least, that's my personal prediction :)

New comment by andyfilms1 in "Microsoft starts canceling Claude Code licenses"

andyfilms1 — Fri, 22 May 2026 17:48:30 +0000

Surely a company as large as Microsoft is actively attempting to build their own models. They couldn't possibly have expected to stake the future of their software development on the conditions of a third party company?

New comment by andyfilms1 in "Google Declaring War on the Web"

andyfilms1 — Thu, 21 May 2026 00:15:35 +0000

It's sustainable in the literal sense, I.E. a tailor can simply tailor forever without needing to constantly worry about keeping up with new tools or technologies, or needing to upgrade or change their methodology constantly.

The tech world is obsessed with moving fast and breaking things, and you can't just do the same thing forever and expect it to always work.

New comment by andyfilms1 in "GenCAD"

andyfilms1 — Mon, 18 May 2026 13:34:21 +0000

A little suggestion in case you decide to try again, just don't worry about constraints. If you're making one-off parts for yourself, just sketch what you need and don't worry about trying to make it parametric. Get the part done and move on.

New comment by andyfilms1 in "GenCAD"

andyfilms1 — Mon, 18 May 2026 13:31:35 +0000

If you use something like OnShape or Fusion, you could easily get comfortable enough to model those parts in about a day. Once get the hang of it, you'll be amazed how fast you can work.

It will take much longer than a day for AI to get to this level, so there's not much to lose by just learning how to use the software now :)

New comment by andyfilms1 in "AI subscriptions are a ticking time bomb for enterprise"

andyfilms1 — Sun, 17 May 2026 14:26:51 +0000

Replacing your workers with AI:

--You lose control over their "salary"

--You lose control over their "schedule"

--Your company becomes reliant on another party that does not share your interests or values, and can stop working for you on a whim for any reason

But AI is definitely good and trade unions are definitely bad, apparently...

New comment by andyfilms1 in "EditLens: Quantifying the extent of AI editing in text (2025)"

andyfilms1 — Thu, 14 May 2026 19:29:55 +0000

I genuinely think any ML approach to detecting ML will always be unreliable. Models can be intentionally poisoned or tricked, and there is a lot of incentive from AI users to do so. It will always be a losing battle against a moving target.

I think in the long run, deterministic algorithmic approaches with complex pipelines will be needed.

New comment by andyfilms1 in "Show HN: AI CAD Harness"

andyfilms1 — Fri, 01 May 2026 21:30:20 +0000

Oh man, a non-web version of this would be an insta-buy from me

New comment by andyfilms1 in "Ti-84 Evo"

andyfilms1 — Fri, 01 May 2026 21:12:34 +0000

Their engineers are still trying to figure out how to make backlit keys. Just give them another two decades, I'm sure they'll crack it.

New comment by andyfilms1 in "To my students"

andyfilms1 — Tue, 28 Apr 2026 01:31:43 +0000

Oh no! Anyway

New comment by andyfilms1 in "The Classic American Diner"

andyfilms1 — Fri, 24 Apr 2026 19:44:02 +0000

It's lunchtime

New comment by andyfilms1 in "I Cancelled Claude: Token Issues, Declining Quality, and Poor Support"

andyfilms1 — Fri, 24 Apr 2026 16:57:49 +0000

Sure, but local AI is still a black box. They can be influenced by training data selection, poisoning, hidden system prompts, etc. That recent Wordpress supply chain hack goes to show that the rug can still be pulled even if the software is FOSS.

New comment by andyfilms1 in "ChatGPT and Codex Down"

andyfilms1 — Mon, 20 Apr 2026 16:34:00 +0000

Sure, but unless you're training them yourself they can still be compromised with poisoning or bias. They're still black boxes even if you're running them locally.