Hacker News: spIrr

New comment by spIrr in "Handbook.md shows that long policy documents do not reliably govern agents"

spIrr — Wed, 29 Jul 2026 13:47:47 +0000

> limited by the number of activated parameters

not sure I got it?

Separately, the frontier labs are kinda pushing us into that behaviour by releasing models with ever-larger context windows.

New comment by spIrr in "Handbook.md shows that long policy documents do not reliably govern agents"

spIrr — Wed, 29 Jul 2026 13:45:57 +0000

As a hobbyist, I find it difficult to figure out how to make Claude stick with some repeating things I want it to do after every major action, like re-evaluate the completeness of tests, update the documentation, etc. And CLAUDE.md/AGENTS.md definitely did NOT help there, sadly.

Handbook.md shows that long policy documents do not reliably govern agents

spIrr — Wed, 29 Jul 2026 13:01:57 +0000

Article URL: https://arxiv.org/abs/2607.25398

Comments URL: https://news.ycombinator.com/item?id=49096969

Points: 279

# Comments: 179

New comment by spIrr in "Vercel's CEO offers to cover expenses of 'Jmail'"

spIrr — Tue, 10 Feb 2026 15:32:47 +0000

Yes, because accounts payable are valued at recognized revenue, and aren't being revalued at cost when written off.

New comment by spIrr in "Ask HN: Is ChatGPT Experiencing a Degradation?"

spIrr — Tue, 09 Dec 2025 12:00:39 +0000

Update: Seems to be an error with 5.1 Pro threads (existing and new inferences). 5.1 Thinking seems to work.

Ask HN: Is ChatGPT Experiencing a Degradation?

spIrr — Tue, 09 Dec 2025 11:51:00 +0000

All my threads have disappeared except the ones inside Projects. They still appear on mobile, but new inference attempts fail with an error. New threads don't get retained either.

Comments URL: https://news.ycombinator.com/item?id=46203884

Points: 2

# Comments: 2

New comment by spIrr in "Search tool that only returns content created before ChatGPT's public release"

spIrr — Mon, 01 Dec 2025 12:06:45 +0000

Thank you. Didn't know that and was, until now, considering paying for a Kagi subscription.

New comment by spIrr in "Busy Bar"

spIrr — Thu, 10 Apr 2025 13:50:29 +0000

This feels like something one could easily procrastinate with (customizable, developer-friendly...).

Juggernaut Pro Flux

spIrr — Fri, 07 Mar 2025 18:42:07 +0000

Article URL: https://runware.ai/blog/juggernaut-pro-flux-the-best-ai-image-generation-model-for-photorealistic-quality

Comments URL: https://news.ycombinator.com/item?id=43292907

Points: 1

# Comments: 0

New comment by spIrr in "iPhones will soon be able to stream live video on 911 calls"

spIrr — Thu, 25 Jul 2024 12:02:27 +0000

I'm curious about how this will work on the receiving end - is that part of the infrastructure also provided by Apple or is there a common capability by 911 dispatch software to support incoming video?

New comment by spIrr in "Datacenter Processors for AI is already 78% of Nvidia's Revenue"

spIrr — Thu, 04 Apr 2024 13:09:42 +0000

When I was a financial auditor, one of the tests on revenue we did was a concentration test, since having a large portion of your revenue coming from a few customers is a risk to business continuity. Wonder how Nvidia folks are thinking of mitigations for that.

New comment by spIrr in "Silencing the Alarm on Aspartame and Cancer"

spIrr — Sun, 06 Aug 2023 13:14:22 +0000

“ Does aspartame cause cancer? There’s a small chance that the answer is yes – but probably only if you consume 50 Diet Cokes every day from birth until death – and even then, we don’t have any evidence that this cancer would lead to premature death. At consumption levels equivalent to the ADIs set by the FDA and other organizations – let alone at lower, more typical intake levels – we have no more reason to suspect aspartame as a carcinogenic threat than we have for suspecting carcinogenicity from chewing with your mouth open or wearing flip-flops in wintertime.”

Silencing the Alarm on Aspartame and Cancer

spIrr — Sun, 06 Aug 2023 13:11:49 +0000

Article URL: https://peterattiamd.com/aspartame-and-cancer/

Comments URL: https://news.ycombinator.com/item?id=37021539

Points: 2

# Comments: 2

Ask HN: What do you think will come after Kubernetes?

spIrr — Fri, 14 May 2021 15:49:06 +0000

I mean, Kubernetes became a thing after Docker became a thing before(and so on, all the way down to VMs), so it's reasonable to assume that something new will eventually emerge and become the industry-standard.

Comments URL: https://news.ycombinator.com/item?id=27155526

Points: 141

# Comments: 170

New comment by spIrr in "Career levels, designation, momentum, etc. (2018)"

spIrr — Sat, 13 Mar 2021 23:21:50 +0000

If you like this, I can only recommend Will Larson’s (the author of the blog post) book called “An Elegant Puzzle: Systems of Engineering Management” published by Stripe Press. It a collection of very tactical things of how to think about engineering careers and managing others.

Axios Wants to Help Companies Write Like Its Reporters–For $10k a Year, or More

spIrr — Thu, 14 Jan 2021 14:25:50 +0000

Article URL: https://www.wsj.com/articles/axios-wants-to-help-companies-write-like-its-reportersfor-10-000-a-year-or-more-11610632800

Comments URL: https://news.ycombinator.com/item?id=25776599

Points: 6

# Comments: 0

New comment by spIrr in "Moral Competence"

spIrr — Tue, 05 Jan 2021 19:03:30 +0000

This is really insightful - thanks for sharing that path! And good for you for not sticking to the sunk cost fallacy through the pivots.

New comment by spIrr in "Google Finance Finally Updated"

spIrr — Wed, 16 Dec 2020 20:03:23 +0000

I would give Koyfin a try - was just recently advertised on a podcast I listen to, and looks interesting.

Ask HN: How do you manage projects in adversarial circumstances?

spIrr — Mon, 16 Nov 2020 14:58:42 +0000

Hey everyone, I've been put in charge to manage a 6-months long project that involves many senior stakeholders and many impacted development teams (i.e., they will have to add new items to their backlogs). Now, I realize that (1) the senior stakeholders have conflicting priorities at their org level (siloed, protecting their roadmaps and capacity), which also seems to trickle down to the working level with all the development teams. In addition, this is also work on top of their regular roadmaps, and there is not really an adequate capacity planning done for this additional effort.

Now, it seems far from the "happy path", but how I do I try to turn this around? What are ways I can still motivate all the impacted parties to take on ownership for the project, bottoms-up?

Any references to research/academia/case studies would be highly appreciated.

Thank you!

Comments URL: https://news.ycombinator.com/item?id=25112389

Points: 3

# Comments: 0

Ask HN: Selling or buying a project? (August 2020)

spIrr — Mon, 03 Aug 2020 16:02:14 +0000

There's a rise of "micro-acquisitions" platforms lately where one can sell their side project, e.g., a bootstrapped SaaS, a mobile app, etc. I figured that the HN community is likely also to have some projects they are considering selling.

Comments URL: https://news.ycombinator.com/item?id=24039437

Points: 3

# Comments: 0