Hacker News: jaggederest

New comment by jaggederest in "My current strategy is to not read any of the code written by my agents"

jaggederest — Mon, 27 Jul 2026 22:22:09 +0000

You just have to make maintainability and readability explicit criteria at the strategic level and hold the AI accountable yourself, same as with testing and linting and the tactical level. Fable can generate code that is absolutely readable by a middle school kid, it'll just be less efficient at it than full jargon-laced fanciness, so finding the sweet spot isn't too hard.

New comment by jaggederest in "My current strategy is to not read any of the code written by my agents"

jaggederest — Mon, 27 Jul 2026 22:19:45 +0000

At the moment I do it manually, but I've experimented with having claude code explicitly call out to codex headlessly for adversarial review, and it works great, I just prefer being able to interact with codex during the review.

If you want even greater fun, launch claude and codex in the same working tree and make them fight it out in real time.

New comment by jaggederest in "Professor's invisible prompt trap catches 32/35 students cheating with AI"

jaggederest — Mon, 27 Jul 2026 21:20:40 +0000

This is going to sound about as dated as "VBScript virus catches 32/35 accounting students cheating with Excel" in a few years, I think.

New comment by jaggederest in "Should you wash your solar panels?"

jaggederest — Mon, 27 Jul 2026 17:06:16 +0000

> I believe that setup circumvents the fee since the only thing using energy is the battery, as though it were an off grid setup.

Entirely based on your jurisdiction, but that kind of zero-export requires the same permits and connection fee as a full net metering in mine. You can't end run the process by just being sneaky, electrical work needs permits and they'll either make you take it out (!!) or bring it into full compliance.

As an example, my utility is one of those ones that requires a separate manual disconnect, even though the vast majority of all-in-one inverters have a built in disconnect. It doesn't make any sense and it's at least two grand in permit + equipment, but that's the rule.

New comment by jaggederest in "I got into YC Startup School by hacking it"

jaggederest — Fri, 24 Jul 2026 18:50:32 +0000

For those like me who didn't understand the reference: https://github.com/garrytan/gbrain

New comment by jaggederest in "Claude Opus 5"

jaggederest — Fri, 24 Jul 2026 18:37:11 +0000

I am very excited for a future where all software is by default modifiable, even shipped binaries, via patches or trampolining, or trickery I don't even know the name of.

New comment by jaggederest in "Micro-SaaS Is Dead. Service With A Software Replaces It"

jaggederest — Fri, 24 Jul 2026 18:35:06 +0000

I actually have a panel that I've collected over time, starting with a former coworker who was visually impaired, and I am now thinking it should be some kind of users-signup-and-give-feedback system, some sort of statistical sampling of a few people across the accessibility needs spectrum.

New comment by jaggederest in "Micro-SaaS Is Dead. Service With A Software Replaces It"

jaggederest — Fri, 24 Jul 2026 17:56:03 +0000

Thus demonstrating that the final accessibility gate must be "the mark I human eyeball". Looks impossible to read without reader mode on my screen.

Your "paper" choice is #bfbfc4 - even going to #ddd makes the APCA go from ~63 to 80 and would probably address the complaints handily. WCAG has a lot of gaps.

New comment by jaggederest in "My security camera shipped a GitHub admin token in its login page"

jaggederest — Fri, 24 Jul 2026 16:14:32 +0000

The cool thing about cryptographic security is that it's an asymmetric amount of cost for both parties - to encode is cheap, to break is many orders of magnitude harder, unlike obfuscation where the difficulty is approximately symmetrical.

New comment by jaggederest in "GigaToken: ~1000x faster Language model tokenization"

jaggederest — Thu, 23 Jul 2026 03:00:08 +0000

Just to clarify, latency is one form of performance, and a separate thing to optimize from total resource usage in more classic "performance critical" situations. That performance might be energy, space, or other dimensions besides latency. It might also be something like reliability, accuracy, precision, or even the very human factors like simplicity, modifiability, and visibility.

Heck, even latency alone you can just reduce the standard deviation and get smoother flows. Little's law is a great callout here too, one of my favorite computer science principles.

New comment by jaggederest in "GigaToken: ~1000x faster Language model tokenization"

jaggederest — Wed, 22 Jul 2026 20:30:17 +0000

To concur it's "latency critical", not "performance critical", people often confuse those two - optimize it all, but especially the chained critical path latency!

New comment by jaggederest in "“We have information that Moonshot distilled Fable for the development of K3”"

jaggederest — Wed, 22 Jul 2026 20:25:17 +0000

> but they are different.

How, and why?

> We could very well end up where content IP is protected, LLM output is not and visa versa with reasonable legal founding, doubtful but plausible.

That is the current state of legal rulings - LLM output is public domain, not copyrightable.

New comment by jaggederest in "“We have information that Moonshot distilled Fable for the development of K3”"

jaggederest — Wed, 22 Jul 2026 19:31:00 +0000

I suspect that, in aggregate, all of the informational output of humanity prior to 2020 has taken more resources to produce than the last few years of LLM research.

New comment by jaggederest in "Jack Dorsey launches Buzz to combine team chat, AI agents and Git hosting"

jaggederest — Wed, 22 Jul 2026 17:54:33 +0000

I am sorry to be the bearer of bad news: That is just massive ACLs and permissions issues with extra steps.

New comment by jaggederest in "Ethics Is the Skill That Matters Most Now"

jaggederest — Wed, 22 Jul 2026 17:53:08 +0000

This kind of god-of-the-gaps thing is kind of amusing to watch, especially since I can clearly see myself doing it over time.

Even if the models never get better than they are now, which is unlikely, they're already equipped to usurp taste, ethics, and everything else to a significant degree. Human feedback will always be valuable because we assign value to human feedback, but if you don't think an ethics-bot will ever be a thing, you're not seeing the current state of things.

New comment by jaggederest in "“We have information that Moonshot distilled Fable for the development of K3”"

jaggederest — Wed, 22 Jul 2026 17:49:44 +0000

Perhaps even more importantly, the current frontier LLM models are self-admittedly the product of enormous quantities of copyright infringement and even less savory inputs, so calling them out for distilling the fruit of that tainted tree reads as highly hypocritical at best.

New comment by jaggederest in "Show HN: HN Hall of Fame – browse 3,100 legendary Hacker News links"

jaggederest — Wed, 22 Jul 2026 17:45:56 +0000

Also notable, the internal list that this may or may not derive from: https://news.ycombinator.com/highlights

New comment by jaggederest in "Jack Dorsey launches Buzz to combine team chat, AI agents and Git hosting"

jaggederest — Tue, 21 Jul 2026 21:03:05 +0000

The issue is when users interact with claude, what can they see? If the AI can see super-secret-payroll-details.md - is it allowed to discuss that with any user? is there some sort of internal ACL within claude that asks pretty please don't tell everyone they're underpaid?

That's why multiuser agents can be an issue.

The other way to do this, is to have claude installed for everyone, and any user's chat with claude uses their personal permissions via something like oauth delegation or AWS IAM role kinda things. But that, again, is complicated ACL logic living through the delegation of roles and access

New comment by jaggederest in "Kimi Work"

jaggederest — Tue, 21 Jul 2026 01:14:12 +0000

I am gently presuming that they are not selling tokens at a loss in their all-you-can-eat plans like the frontier models are, but maybe I'm wrong.

New comment by jaggederest in "Kimi Work"

jaggederest — Mon, 20 Jul 2026 21:12:03 +0000

I am confused why they don't just offer a subscription with bring-your-own-harness, that would be a differentiator from claude/codex I think.