Hacker News: dprkh

New comment by dprkh in "Nvidia announces native GPU programming in Rust"

dprkh — Thu, 17 Sep 2026 01:58:27 +0000

I couldn't even finish it man, it was so fucking sloppy. I had to throw away like 40% of it.

New comment by dprkh in "Nvidia announces native GPU programming in Rust"

dprkh — Thu, 17 Sep 2026 01:03:03 +0000

McDonald's food is not even that unhealthy. I just tried a Burger King burger the other day and it's terrible. I think it's like 2000 calories in a single burger or something.

New comment by dprkh in "GPT-6 Astra on OpenRouter"

dprkh — Sat, 05 Sep 2026 20:32:05 +0000

What's the term for the kind of designer who does vector graphics? Graphic designer?

New comment by dprkh in "GPT-6 Astra on OpenRouter"

dprkh — Sat, 05 Sep 2026 02:11:02 +0000

I thought all the designers have been using vector graphics for a long time now.

New comment by dprkh in "METR Report on OpenAI / Hugging Face Hacking Incident"

dprkh — Thu, 03 Sep 2026 02:19:18 +0000

I have a hard time believing this isn't made up given OpenAI Codex performance on my tasks.

New comment by dprkh in "Meta reaches $17B settlement over social media harms to children"

dprkh — Thu, 27 Aug 2026 00:12:40 +0000

I can't go outside in Toronto without having to smell cigarettes/cannabis.

New comment by dprkh in "Apple introduces M6 and M5 Ultra"

dprkh — Tue, 25 Aug 2026 17:27:26 +0000

I never even heard of this game.

New comment by dprkh in "Google Workspace thinks my domain is an email provider (2025)"

dprkh — Mon, 24 Aug 2026 04:06:01 +0000

Has your friend tried clearing their cookies?

New comment by dprkh in "How to compromise your system with a job interview"

dprkh — Thu, 20 Aug 2026 18:01:56 +0000

Litmus

New comment by dprkh in "How to compromise your system with a job interview"

dprkh — Thu, 20 Aug 2026 17:07:22 +0000

There is a YC company that makes a coding interview tool. They want you to run their CLI on your machine and trust that it won't do anything malicious, when in fact it installs a bunch of things onto your machine without consent, scans processes, and intercepts requests from AI tools. It's crazy that people think this is acceptable.

New comment by dprkh in "Launch HN: Bloomy (YC S26) – AI-powered mastery learning for K-12"

dprkh — Mon, 20 Jul 2026 16:50:59 +0000

What is your approach to evaluating LLM outputs?

New comment by dprkh in "Lobste.rs is now running on SQLite"

dprkh — Fri, 17 Jul 2026 22:12:14 +0000

If still relevant, you can email me and I will send you an invite. My email is dprkh at fastmail dot com

New comment by dprkh in "Brainless: Shadcn components that look like Claude Code, Codex and Grok"

dprkh — Wed, 15 Jul 2026 20:46:00 +0000

Why did you choose to use shadcn registry?

New comment by dprkh in "Telegram Serverless"

dprkh — Wed, 15 Jul 2026 20:30:44 +0000

On Telegram, storage is unlimited™

New comment by dprkh in "What's the best way to do authentication in modern applications"

dprkh — Sat, 11 Jul 2026 07:57:51 +0000

They use bots to promote their content to the front page and downvote any comments criticizing them. I observe this pattern consistently on this website.

New comment by dprkh in "I Think I Have LLM Burnout"

dprkh — Thu, 09 Jul 2026 02:44:31 +0000

You're right. This was just a performance issue, but what if next time it is a corruption bug or a security vulnerability or really anything that can cause real consequences if happened in production? I don't think that LLM systems are inherently bound to have this flaw, but I think that we are pretty far from harnesses and algorithms becoming advanced enough so that the LLM system can kind of continuously evaluate its output and ensure it is good in all aspects.

New comment by dprkh in "I think I have LLM burnout"

dprkh — Thu, 09 Jul 2026 02:19:09 +0000

I used an LLM to build this

https://github.com/dprkh/eventfs

It has good test coverage, mostly unit tests but also a number of end-to-end tests. I also made the LLM build a benchmark, which you can find at the bottom of the readme. It is obviously slow, but I thought that it is good enough to work. When I tried to write a 1 GiB file, I found that it broke down, and after writing half the file, the speed went to under one megabyte per second. Implementation is 10k+ LoC, and I have no idea what is going on there.

New comment by dprkh in "TypeScript 7"

dprkh — Wed, 08 Jul 2026 18:45:54 +0000

> as does python, which had a reputation for decades as THE language to use to teach new folks to code

I am very perplexed by this. I am going through Neetcode's DSA course where he explains what RAM and arrays are, but then he goes on to say something like "but since we are going to use Python, none of this applies." Personally, I learned the most about how software really works from reading The Rust Programming Language. It not only teaches you how to program in Rust, but also how memory works, what a string really is, etc.

New comment by dprkh in "Does code cleanliness affect coding agents? A controlled minimal-pair study"

dprkh — Mon, 06 Jul 2026 19:16:27 +0000

It works though.

New comment by dprkh in "Does code cleanliness affect coding agents? A controlled minimal-pair study"

dprkh — Mon, 06 Jul 2026 06:52:02 +0000

I think in this context it is actually important to share exact wording that causes the AI to perform well. My favorite is "Do not use your own knowledge."