Hacker News: Robdel12

New comment by Robdel12 in "Oceans hit highest temperature on record"

Robdel12 — Mon, 24 Aug 2026 22:25:11 +0000

I always will think of how everyone laughed at Bernie Sanders' reply in ~2016 to "What is the greatest threat to national security" and he replied "Climate change".

It's going to get ugly when large swaths of the human population struggles to get food & clean water.

New comment by Robdel12 in "U.S. Department of Energy Launches the Genesis Open Models Initiative"

Robdel12 — Sat, 08 Aug 2026 12:07:36 +0000

Poolside.ai as well

New comment by Robdel12 in "Almost no skill required to cook a steak"

Robdel12 — Thu, 06 Aug 2026 19:03:20 +0000

https://www.youtube.com/watch?v=rDBNVO7K0kA this is a really good little episode from PBS

New comment by Robdel12 in "JetZero"

Robdel12 — Sun, 26 Jul 2026 04:10:51 +0000

This reeks of Vern Raburn and Eclipse Aviation all over again, haha. That failed terribly. SV’s ego is pretty funny sometimes. Eclipse even tried to develop their own avionics.

New comment by Robdel12 in "Nvidia, Microsoft, Meta warn against overregulating open-weight models"

Robdel12 — Fri, 24 Jul 2026 21:58:09 +0000

https://www.anthropic.com/news/donation-public-first-action

Probably because anthropic is pouring $40 million dollars into a political pact to regulate models. And they have not been quiet about wanting to ban/regulate OSS models either.

I have no idea why HN still treats them like ~their~ they are the ethical good guy.

Edit: I can’t English. At least we know this came from my dumb swishy brain

New comment by Robdel12 in "Why Software Factories Fail (or: harness engineering is not enough)"

Robdel12 — Thu, 23 Jul 2026 19:40:49 +0000

I 100% whole heartily agree with this. Anyone saying the models made a huge leap in fall 2025 / spring 2026 still aren't looking at whats going on. Just this weekend I thought I had a solid plan and evals/tests to let 5.6 sol work unsupervised over night while I slept and it made an absolute mess. Somewhere in the loop it had to make a decision and it made a wrong one. Making everything beyond that trash. The code 'worked' but it had the wrong system design and wrote the most brittle tests around its assumption, validating its own decision. It turned into a nasty feedback loop for the model. This is 5.6 sol high.

The models write good/great code. I'm very happy to never write code again but models are no where near good enough to run off on their own without a human in the loop. Models LOVE to cheat. Just peek at the tests they write.

And before you come at me, I have built 7 products in the past 1.5/2 years with agentic engineering, all with users. One of those projects is _dead_ because I let the vibe go too hard at the same time Anthropic decided to nerf both their harness and their models silently. If you care to look at the source: https://github.com/Robdel12/OrbitDock I spent a week or so and like a billion+ tokens trying to refactor and save it. It just wasn't worth it.

I wish people would be pragmatic about this. I get the dream is to let it do everything and not to be in the loop, because being in the loop is exhausting. But if you want to make whatever you're building be robust and survive more than 6 months, you have to. I don't care how good your tests, plans, skills, etc are. At some point the model will have to decide something and it'll be the wrong one. Compounding the slop from there forward.

New comment by Robdel12 in "OpenAI and Hugging Face address security incident during model evaluation"

Robdel12 — Tue, 21 Jul 2026 23:23:57 +0000

> and an even more capable pre-release model, all with reduced cyber refusals for evaluation purposes — while being internally tested on a benchmark of cyber capabilities.

This is pretty wild but also I think this is doing a lot of heavy lifting here. This was not a model everyone has access to. I mean, still insane.

New comment by Robdel12 in "The infinite scroll may become endangered if controversial Calif. law passes"

Robdel12 — Mon, 13 Jul 2026 23:58:54 +0000

I’m going to be honest, this kind of regulation would make me disable the site(s) for the state (if there are fines, etc). I don’t have the time to play these games for my tiny projects

New comment by Robdel12 in "Apple sues OpenAI, accuses ex-employees of stealing trade secrets"

Robdel12 — Fri, 10 Jul 2026 23:07:53 +0000

OpenAI is about to get ROCKED on this. From this report, this looks open and shut. Apple has basically infinite money and incredible lawyers. Not sure what OpenAI can counter with unless they have clear, hard evidence this hasn’t been happening.

New comment by Robdel12 in "Leaking YouTube creators' private videos"

Robdel12 — Sun, 05 Jul 2026 04:58:09 +0000

You compared trains to YouTube videos, I cannot take _you_ seriously.

New comment by Robdel12 in "New serious vulnerabilities spiked around release of Claude Mythos Preview"

Robdel12 — Sat, 04 Jul 2026 02:52:38 +0000

…are we really drawing conclusions on this starting at April? When it was released in June?

New comment by Robdel12 in "Claude Sonnet 5"

Robdel12 — Wed, 01 Jul 2026 01:57:16 +0000

Dario’s own mouth https://x.com/coinbureau/status/2071330294452666695/mediavie...

New comment by Robdel12 in "Salesforce to Acquire Fin (formerly Intercom) for $3.6BN"

Robdel12 — Mon, 15 Jun 2026 13:28:06 +0000

Huge congrats to the intercom team on that! Intercom was pretty great for me during my Percy support years. Their current AI direction is rough, though. When I went to use them for my product I was (am?) building, I found intercom unrecognizable with the entire ai pivot. Honestly made me sad but that’s how it is these days

New comment by Robdel12 in "Amazon Says Its Data Centers Use 2.5B Gallons of Water"

Robdel12 — Thu, 11 Jun 2026 15:37:12 +0000

You find growing an incredibly water heavy crop in a place that doesn’t have the water supply to do so a weird comparison? And it’s not even a food that’s needed, so you can’t stand on that either.

New comment by Robdel12 in "The ways we contain Claude across products"

Robdel12 — Thu, 04 Jun 2026 12:57:19 +0000

Hahahaha yeah, makes sense. Every time I use claude after codex I feel like I’m holding Claude’s hand the entire time. I imagine they have a lot of containing they have to do internally

New comment by Robdel12 in "U.S. Army Corps of Engineers Bay Model"

Robdel12 — Thu, 04 Jun 2026 12:55:13 +0000

This is neat to see. US army crops of engineers is a negative “word” to me after growing up in FL and they destroyed so many ecosystems. And the entire Everglades. They’re still at it now. My family has basically spent the past 30 years fighting a ware they put in on our natural creek. It killed the creek, it shrunk the flow to the size of the culvert.

So, It’s neat to see something competent! Imagine if they modeled what cutting off the natural draining to the Everglades would do :p

New comment by Robdel12 in "Anthropic confidentially submits draft S-1 to the SEC"

Robdel12 — Mon, 01 Jun 2026 18:51:57 +0000

> Compounding the problem, labs in China often release dual-use capable models as open-weight. Once a model is open-weight, safeguards that do exist can be removed, making the model available to any state or non-state actor to use for malicious purposes, including the cyber and CBRN misuse those safeguards were built to prevent.

https://www.anthropic.com/research/2028-ai-leadership

They are already starting that now.

New comment by Robdel12 in "Dynamic Workflows in Claude Code"

Robdel12 — Thu, 28 May 2026 19:25:11 +0000

> Rewriting Bun with dynamic workflows

There ya go, the rewrite was for marketing.

New comment by Robdel12 in "Incident with Pull Requests, Issues, Git Operations and API Requests"

Robdel12 — Wed, 27 May 2026 18:05:20 +0000

I know it’s super fun to shit on GitHub and everyone’s favorite thing to say is “build a competitor!”

They’re trying to scale from 1 billion commits last year to over 14 billion this year. I have zero desire to try and manage that scaling. Basically being DDOS’d by agents all day now.

https://x.com/kdaigle/status/2040164759836778878

New comment by Robdel12 in "Bun support is now limited and deprecated"

Robdel12 — Fri, 22 May 2026 20:12:25 +0000

I have no idea how that’s what you get from this. I don’t want my project using any tech that decides to take 6 days to rewrite the entire library with AI. That is at its core an engineering decision.

No healthy engineering team is going to do that. And I’d want to distance myself as far as I could from a project that behaves like that.