Hacker News: saadn92

New comment by saadn92 in ""People who don't use AI will be left behind""

saadn92 — Wed, 29 Apr 2026 19:53:12 +0000

People also used to say that Google or calculators will make you dumber. Neither happened. Won't happen with this either.

New comment by saadn92 in "Ghostty is leaving GitHub"

saadn92 — Tue, 28 Apr 2026 20:08:07 +0000

Nothing stupid about caring deeply about tools that shaped your career. GitHub wasn't just a SaaS for a lot of us it was where we learned to build. The fact that you're emotional about it says more about how much you gave to that platform than anything else.

Ghostty will be fine wherever it lives because people follow the project and not where it's hosted. Best of luck!

New comment by saadn92 in "Who owns the code Claude Code wrote?"

saadn92 — Tue, 28 Apr 2026 19:05:32 +0000

I agree with this sentiment, because the person directing the agent can still direct it in a way where it'll produce a better or worse output than another person directing it.

New comment by saadn92 in "To my students"

saadn92 — Tue, 28 Apr 2026 00:52:03 +0000

What gets me is the craft point. I've shipped more useful software in the last year than probably the previous five combined, and most of that is because I stopped treating code as the artifact and started treating the product as the artifact. The craft moved up a layer.

> until it is clear and elegant

New grads who spend weeks refactoring code are going to get lapped by new grads who ship something and iterate. There's just a faster feedback loop now.

New comment by saadn92 in "NPM Website Is Down"

saadn92 — Mon, 27 Apr 2026 21:53:30 +0000

ha, github is down too

New comment by saadn92 in "4TB of voice samples just stolen from 40k AI contractors at Mercor"

saadn92 — Mon, 27 Apr 2026 16:41:36 +0000

The irony runs deeper than the free analysis offer. The whole Mercor contractor relationship was this exact pattern: hand over studio-quality voice recordings and ID scans to get paid for data labeling work that didn't require either. "Explicit consent" was buried in the terms, and people clicked through because they needed the paycheck.

Now 40k people have learned that biometrics aren't passwords. You can't rotate your voice.

New comment by saadn92 in "Microsoft and OpenAI end their exclusive and revenue-sharing deal"

saadn92 — Mon, 27 Apr 2026 16:18:49 +0000

That's a pretty good swap if you're Microsoft. Exclusivity was already unenforceable in practice, and they were going to have to either sue their biggest AI partner or let it slide. Instead they got the agi escape hatch closed and a revenue cap that at least makes the payments predictable

New comment by saadn92 in "Pgbackrest is no longer being maintained"

saadn92 — Mon, 27 Apr 2026 15:34:20 +0000

[flagged]

New comment by saadn92 in "AI should elevate your thinking, not replace it"

saadn92 — Sun, 26 Apr 2026 20:12:43 +0000

Hard disagree. I feel like I'm thinking a lot more now because I have so many parallel projects going on at the same time. AI has allowed me to really, truly create in a way that I've never done before. Yes, my coding skills probably aren't as sharp as they used to be, but my system design skills are at an all time high. Don't blame the tool.

New comment by saadn92 in "Why has there been so little progress on Alzheimer's disease?"

saadn92 — Sun, 26 Apr 2026 14:34:04 +0000

This is similar to what we see in software architecture. There's a team that picks a framework or pattern early, then builds everything on top of it, and by the time evidence shows up that the foundation was wrong. Now, switching costs are so high that it's cheaper to keep building on the broken foundation than to start over. The amyloid hypothesis reminds me of technical debt. The "cabal" wasn't conspiring, but they were just rationally protecting their sunk costs, same as any engineering org that can't migrate off a bad database choice.

New comment by saadn92 in "Asahi Linux Progress Linux 7.0"

saadn92 — Sun, 26 Apr 2026 14:23:24 +0000

The real answer is probably simpler than anyone here is making it. Apple hardware margins are healthy enough that selling macbooks to linux users is pure profit, so no services lock-in needed. However, the moment they officially acknowledge Linux support, then it becomes a support surface. Every kernel panic becomes a genius bar visit. Every driver bug becomes a tweet at @AppleSupport. It's the value of plausible deniability. The Asahi team being unofficial is actually the best possible outcome for Apple in that they get hardware sales to Linux enthusiasts without any support burden.

New comment by saadn92 in "Using coding assistance tools to revive projects you never were going to finish"

saadn92 — Sun, 26 Apr 2026 03:18:52 +0000

It's crazy. I have something like 120 personal tools at this point and the pattern you describe is exactly right. The bottleneck moved from implementation to context switching. I started keeping a markdown file at the root of every project that captures state and next steps whenever I stop working on it, purely so I can resume without the 20-minute "wait where was I" tax.

There's just no pressure to handle edge cases or write docs for people who'll never use it. Just solve exactly your problem and move on.

New comment by saadn92 in "1-Bit Hokusai's "The Great Wave" (2023)"

saadn92 — Sat, 25 Apr 2026 16:58:38 +0000

This is what keeps me coming back to HN. Someone spent years recreating woodcut prints pixel by pixel on a quadra 700 using aldus superpaint at 512x342. I feel like the constraint is what caused it to be. The 1-bit forces you to solve every gradient and texture with pure composition, which means you can't cheat with color or resolution. I forgot who said it, but constraints breed creativity.

New comment by saadn92 in "Plain text has been around for decades and it’s here to stay"

saadn92 — Sat, 25 Apr 2026 16:55:23 +0000

Not just TUIs the whole stack is converging back to text. I run ~15 personal tools and every one that survived past the first month stores data as JSON/markdown in git repos.

Text in git gives you versioning, sync, grep, and you can hand the whole thing to an LLM with zero serialization. It's perfect for me.

New comment by saadn92 in "Show HN: A Karpathy-style LLM wiki your agents maintain (Markdown and Git)"

saadn92 — Sat, 25 Apr 2026 16:51:32 +0000

I've been running a variation of this for ~6 months. What seems to work: a background process that reads conversation transcripts after sessions end and then extracts decisions/rejected approaches into structured markdown. I review before I promote it into the context.

Phantom: Web Automation Without a Browser

saadn92 — Fri, 24 Apr 2026 16:08:21 +0000

Article URL: https://saadnaveed.com/writing/phantom-web-automation-without-a-browser/

Comments URL: https://news.ycombinator.com/item?id=47892111

Points: 2

# Comments: 0

New comment by saadn92 in "Bitwarden CLI compromised in ongoing Checkmarx supply chain campaign"

saadn92 — Fri, 24 Apr 2026 12:06:42 +0000

If anyone is looking for something local: https://github.com/saadnvd1/xpass

New comment by saadn92 in "An update on recent Claude Code quality reports"

saadn92 — Thu, 23 Apr 2026 20:24:22 +0000

No, I wouldn't. I'd like some transparency at least.

New comment by saadn92 in "An update on recent Claude Code quality reports"

saadn92 — Thu, 23 Apr 2026 19:40:21 +0000

I leave sessions idle for hours constantly - that's my primary workflow. If resuming a 900k context session eats my rate limit, fine, show me the cost and let me decide whether to /clear or push through. You already show a banner suggesting /clear at high context - just do the same thing here instead of silently lobotomizing the model.

New comment by saadn92 in "Show HN: Hydra – Never stop coding when your AI CLI hits a rate limit"

saadn92 — Wed, 22 Apr 2026 17:35:34 +0000

thank you!