gravypod

11 ms·

1.
by gravypod 13d ago
> For the many people who resist the AGI label possibly ever being achieved, I'd be curious to hear takes on what would make you think Astra is yet to be AGI, and what would still need to be achieved for this to effectively be AGI f
2.
by gravypod 13d ago
My success rate for solving software engineering challenges encountered in my day jobs has been near 100% for my entire career. I can only think of a few tasks I kicked back and said they were impossible. For example, after trying to get a
3.
by gravypod 18d ago
> 1. There aren't enough humans in OpenAI to "peak at the output tokens during the run" of every AI agent. For a training run, you will often do this. You'll randomly sample some of the forward pass. You can also imag
4.
by gravypod 20d ago
It's a shame because I like GitLabs UX and runners much more than other forges. I've been considering forking gitlab and removing complexity. GitLab uses multiple GB of memory idle with no users. It's sad.
5.
by gravypod 21d ago
Why are these models able to reduce parameters but keep quality? I know the original intuition was scale data + params = quality but it looks like we have hit an s curve on improvements from pure scaling? Is this just because we are in a me
6.
by gravypod 22d ago
People are discovering go links from first principles.
7.
by gravypod 23d ago
Is there a market for an actual GitHub competitor now that everyone is looking around and is very angry about stability?
8.
by gravypod 26d ago
The linked paper is a much more interesting read: https://arxiv.org/pdf/2608.16157 I think there are many other significant inference improvements which an be built out that are being ignored because the API between cl
9.
by gravypod 26d ago
I recently built a piece of code which downloads a bulk set of data, indexes it for search, and then serves a pretty web UI on top of this with the help of some AIs. Normally I would have reached for sstables, sqlite, etc. This time, becaus
10.
by gravypod 27d ago
> If they are truly rare, then they are likely not valuable, otherwise there would be more copies or their contents could be found elsewhere. That's what I keep saying about the van Goghs I burn to heat my home but everyone is still
11.
by gravypod 1mo ago
We will not reverse climate change by consuming less. The solutions are equally political and technological. Driving a smaller vehicle does help but real solutions would be: building homes next to jobs, public transportation, electrificatio
12.
by gravypod 1mo ago
I have taken a few cars on that drive. Adding in traffic, I usually have to fill up at the gas station just outside of Boston that has the burger king in it. In my Mirage can get me to Boston, drive around a bit, and usually just need to fi
13.
by gravypod 1mo ago
Mitsubishi just discontinued the car line that I drive [0]. The Mirage was great. I can drive from NYC to Boston on ~10 gallon. The only downsides were: No arm wrest, no sunglasses holder, and the media console won't turn on if you lea
14.
by gravypod 2mo ago
If they did have a super powerful secret AI why would avoid: 1. Selling access to the US? 2. Ship more and better software? 3. Talk about it publicly? If you had a secret AI better than anything currently available the mere mention of this
15.
by gravypod 2mo ago
What would the benefit of this be? If China stops open weights, US labs still have the intelligence frontier. Maybe once Chinese models have speed, cost, and intelligence beat but right now they don't. Undermining out entire economy by
16.
by gravypod 2mo ago
You would need to demonstrate an impact for a line item that big. A 3k person org at $200/seat would be >$7mill/year. That places you very well into self hosting Kimi K3 territory and the never having: price hikes, dependency o
17.
by gravypod 2mo ago
> With Zig you usually only need the Zig toolchain, which runs just fine in a regular Windows cmd.exe. It's amazing how complicated we made things with bad software engineering and how good we can make things with good software engi
18.
by gravypod 2mo ago
> And second, I am completely uninterested in what AI has to offer. To me (and most normal people) knowledge pursuit is about inspiring others with discovery and the intellectual joy of doing something yourself. I often hear that people
19.
by gravypod 2mo ago
Yes, exactly. The idea was that people could be content with our system if people's material conditions were good. We seem to have lost this insight and continued for the next 50-60 years to degrade the quality of life.
20.
by gravypod 2mo ago
"No man who owns his own house and lot can be a Communist. He has too much to do." - William Levit (the Levittown one)
21.
by gravypod 2mo ago
This is not always a great idea. Especially if you have a water proofing membrane that allows drying out. You never want to trap moisture in a cavity. You want to be able to dry into the wet side.
22.
by gravypod 2mo ago
Neither is grout. Grout is water permiable. You shower is water proof from the water proof backing we install the tile on top of. Cement board + roll on or something like kerdi system. https://youtu.be/B5DVpruLygw?si=jD8eMHH
23.
by gravypod 2mo ago
How is this more expensive than a Stream Deck? Shouldn't OpenAI be able to undercut existing software / hardware stacks due to how automated their engineering team is? https://openai.com/supply/co-lab/wor
24.
by gravypod 2mo ago
When people ask for a commercial service hookup there is usually a charge to bring transformers and service lines to the building. This is a constant cost based on the load. The infrastructure to run the service lines out to a new commercia
25.
by gravypod 2mo ago
Forcing all costs into externalities is also not ideal. Right now, some state's systems, force residential consumers to pay for new infrastructure for new demand while only minimal costs are paid by the new customer.
26.
by gravypod 2mo ago
What kind of harness does the exploration? Where did the corpus of Lean proofs come from? Is the code backing Ton 618 open source?
27.
by gravypod 2mo ago
I've found a good balance in looking through sessions based on user feedback (ex: user thumbs down and submits it into a review queue) to build an understanding of usage patterns and failure modes. This is very boring but critical to u
28.
by gravypod 2mo ago
I think part of this is the organization of the training. Cuba trains doctors for likely much less than $500k USD. Also, the training we give to a US solider is very expensive but we have "productionized" the process. The people w
29.
by gravypod 2mo ago
There was an article posted on HN recently about the asymmetry in cost of providing an ambulance service vs the cost of the per-ride service. The cost of a medical degree, and the training on top of the degrees, may seem waaay too high but
30.
by gravypod 2mo ago
Medicine and jokes

More ›