weitendorf

11 ms·

91.
by weitendorf 4mo ago
You should note that Claude Design is most likely a DPO->PPO->Actor-Critic bootstrap play: https://arxiv.org/abs/2305.18290 / https://en.wikipedia.org/wiki/Proximal_policy_optimization
92.
by weitendorf 4mo ago
It’s not cheaper to run Claude in your own GPUs rather than the $200/mo for certain workloads. For a large portion of what I work on, the bottleneck is my time, not tokens. You certainly could throw more tokens at it but if you need
93.
by weitendorf 4mo ago
This is what I’ve been doing. I’m not even against external funding, I just see it as instrumental to the ultimate goal of building a sustainable business. Venture capital is basically a super high interest loan so it’s only something you w
94.
by weitendorf 4mo ago
There are basically two tiers of "Chinese models" in this context, the "edge" sized ones with ~30B parameters or less, and the big ~1T models that can basically only run in the datacenter. I don't think it's as
95.
by weitendorf 4mo ago
It's through my startup, so both I guess. Generally I find my bottleneck to be attention and focus, and the opportunity cost of not going back to work at my prior employers absolutely dwarfs the amount of money I spend on tools, so it&
96.
by weitendorf 4mo ago
I pretty strongly feel the opposite way. Granted I have not used deepseek enough to “know” their model idiosyncrasies as well as Anthropic, so there is a partial skill issue. But I just find it really hard to justify using a less powerful m
97.
by weitendorf 5mo ago
Working on it, they all already expose a tool interface, you just have to know where to look for it and how to use it!
98.
by weitendorf 5mo ago
A formative moment for me was reading Richard Stallman's writing on the GNU website and seeing him quote [0] Rabbi Hillel [1]: "If I am not for myself, who will be for me? If I am only for myself, what am I? And if not now, when?&
99.
by weitendorf 5mo ago
It would require much more than a couple of queries per day, I want to basically do bulk ingestion and search/evaluation/integration across tens of thousands of videos and software projects (if it were cheap enough and smart enoug
100.
by weitendorf 5mo ago
I think there is a reasonable basis for taking a gamble that small models capable of fitting on a 32GB card will continue to advance over the next 5 years and eventually approach Gemini Flash 3.5 / Sonnet 4.6 levels of capabilities, wh
101.
by weitendorf 5mo ago
In the long run cloud gaming is inevitable, it’s just more economically efficient for the cost of the hardware required to render graphics to be amortized across consumers and not sit idle when being unused by collocating them with game ass
102.
by weitendorf 5mo ago
TSMC doesn’t get to take the profit that currently accrues to Nvidia and Apple, even though they absolutely could from a business/leverage perspective, because they are an economic colony of the United states and hiking their prices (w
103.
by weitendorf 5mo ago
If you factor in Nvidia’s profit margin due to the scarcity of the current bleeding-edge chips there is a path to a much larger cost reduction still. There’s a lot to criticize Sam Altman for saying or popularizing culturally but I’ve come
104.
by weitendorf 5mo ago
That's awesome, I'll definitely try it now. To be clear my thought process was that I couldn't find any info about you or any human accountable for the project from its main site, and IME that's a strong yellow/red
105.
by weitendorf 5mo ago
This looks really cool! Unfortunately since it's not FOSS and there's no information about the company/individuals behind it, or even a way to pay for it/get licensing information from your UI, there is absolutely no way
106.
by weitendorf 5mo ago
Dude it's literally on bloop. It's a bloop original
107.
by weitendorf 5mo ago
With 4.5, I think because I would prompt it/guide it towards an outcome by calling it “the dream: <code example>” it would get almost reverential / shocked with awe as it got closer to getting it working or when it finally p
108.
by weitendorf 5mo ago
Completely agree, top down “alignment” and RLHF is actually quite primitive and uses a lot fancy words to describe what is essentially just hitting the machine with a stick without the nuance, context, or feedback to help it model why the f
109.
by weitendorf 5mo ago
It actually probably wouldn’t be too expensive or difficult to finetune those sayings out of default behavior if it were made accessible to you, you could even automate most of the relabeling by having the model come up with a list of idiom
110.
by weitendorf 5mo ago
I think if you see it as weird social phases that the model lacks the self-awareness to identify as kinda embarrassing, it makes more sense. Like if a human were going around saying “for the culture!” so much at work that they didn’t realiz
111.
by weitendorf 5mo ago
Good catch —- even though the prompt explicitly forbade training on user data, a couple of gremlins in the pretraining pipeline disabled the sample filtering during test runs so that remove_the_gremlins.sh would only run on commit, not duri
112.
by weitendorf 6mo ago
Get outta my swamp! Just kidding, it’s cool to see other people working on this stuff. I think right now this is still a bit too fresh out of Claude Code to be usable by anybody but the people developing it. I got to around the same point w
113.
by weitendorf 6mo ago
The UIs all bake in system prompts and other tunable configs that the API leaves open, so does Claude Code and other harnesses. So anything you notice different over the API when you're controlling the client is almost certainly that.
114.
by weitendorf 6mo ago
I find the most value to be in eval loops and multi-agent setups where a specialized or cheap model gets tasks that take load off the smarter model. Most of the value in agentic development IMO is in the feedback loop/ability for the m
115.
by weitendorf 6mo ago
It’s to stop you from getting RL traces or using Claude without paying the big bucks for the Enterprise Security version I really like Anthropic models and the company mission but I personally believe this is anticompetitive, or at least, a
116.
by weitendorf 6mo ago
They are training them on decompilation and reverse engineering/blackbox reimplementations/pentesting because it’s one of the best ways to generate interesting and rare RL traces for agentic coding AND teach them how lots of thing
117.
by weitendorf 6mo ago
This is a price discrimination/upsell strategy. Sure, if you just want software, use our public model. Don’t worry; it’s safe. But if you want your model to be secure , and you want to deal with dangerous stuff, contact us for prici
118.
by weitendorf 6mo ago
Hey OP, sorry for the negativity, I think most of these commenters right now are pretty off-base. My company is building a lot of API infrastructure and I thought this was a great write up!
119.
by weitendorf 6mo ago
Hey, I've been getting into visual processing lately and we just started working on an offline wrapper for Apple's vision/other ML libraries via CLI: https://github.com/accretional/macos-vision . You can
120.
by weitendorf 6mo ago
Guys, I found out about this technology called Cascading Style Sheets recently and I think it's the missing piece we've been looking for. It lets you declaratively specify layout in a composable, hierarchical system based on somet

More ›