iwontberude

7 ms·

91.
by iwontberude 4mo ago
Yes, they are typically pretty useful for reproducing the setup. I argue also they are superior to most human made README.md because they explain what the project is at an abstract level, use more features of GitHub Markdown to communicate,
92.
by iwontberude 4mo ago
People are wasting tokens with perfectly fine models by not planning effectively. Anyone who says Opus is not capable of doing what Fable can probably are not planning effectively and just praying to the one shot gods.
93.
by iwontberude 4mo ago
It reads like satire
94.
by iwontberude 4mo ago
If you knew about how much man power it takes to maintain, evaluate and improve agentic workflows, I don’t think you would write such a thing. In this context, AI is a jobs program for permanent employment.
95.
by iwontberude 4mo ago
If it outputs copyrighted material, which it does handily, then it doesn’t really matter.
96.
by iwontberude 4mo ago
It seems like the PM and SWE impetus is at odds
97.
by iwontberude 4mo ago
Your models aren’t big enough and they are forgetting about the tools. Try a larger model. If you can’t, then your rig was too underpowered anyways.
98.
by iwontberude 4mo ago
Because your rights (including property rights) are only what other people will respect, there is no true god head that swoops down to ensure it.
99.
by iwontberude 4mo ago
They are notoriously stingy and underpay.
100.
by iwontberude 4mo ago
I pray you are right, this country deserves a good spanking.
101.
by iwontberude 4mo ago
They are good if you were clever enough to buy a powerful enough rig before memory went up. For everyone else I say just wait. M1 Ultra 128GB and higher is sufficient to run gemma4:31b-mlx or qwen3.6:35b-mlx with subagents. It’s only slow i
102.
by iwontberude 4mo ago
Not to be crass buts it’s basically slave labor with free lunch
103.
by iwontberude 4mo ago
Unless you were black then you might think good god these yts gonna displace us again with a new new deal.
104.
by iwontberude 4mo ago
He ruined Google, started by shattering the culture into a million pieces. He has negative charisma and no imagination. He was best as a mid level manager.
105.
by iwontberude 4mo ago
Luckily for programmatic or logic following, smaller models tend to do better, it can be surprising at first to see the more expensive models do worse at a task but it’s true.
106.
by iwontberude 4mo ago
I’ll take the under on that bet any day. Maybe around 75% is where I would feel more confident. 25% of a country is more than enough to meaningfully shift perception.
107.
by iwontberude 4mo ago
A distilled model for specific domains is both nearly free and commodity. There is no money in AI outsourcing businesses at scale. If anything you are just a cost center that companies want to squish because your value a negated by the amou
108.
by iwontberude 4mo ago
As long as you use models trained or distilled for your use case, there is no need to waste compute on trillions of parameters. Anthropic and OpenAI are proving that the “everything” models are not a sustainable business model. Just-In-Time
109.
by iwontberude 4mo ago
Nvidia wants to create a digital twin of the real world that military can use to plan their next military operations. All of those scans of cities and inside buildings will be a virtual world where war makers can plan to a very fine detail
110.
by iwontberude 4mo ago
These fools are compromised by nation states and the data isn’t just in their hands. This is why you shouldn’t collect certain types of data.
111.
by iwontberude 4mo ago
Yeah they just exfiltrated it for free.
112.
by iwontberude 4mo ago
Hopefully they will let us live comfortably like pets as sterile creatures that will slowly die out and not be replaced. This could limit human misery in the future.
113.
by iwontberude 4mo ago
It’s not unnecessary. If anything it seems necessary.
114.
by iwontberude 4mo ago
It doesn’t get bored or demotivated, but it also lacks interest and motivation generally so it comes with the same pitfalls of having nothing to lose and being utterly unaccountable, (e.g. destructive actions, lying, and being coercive or M
115.
by iwontberude 4mo ago
Wow what an insightful comment we haven’t heard repeated for over a decade here.
116.
by iwontberude 4mo ago
Just need a decent Mac Studio and they are plentiful in used condition and affordable.
117.
by iwontberude 4mo ago
Keep working on your agentfu because there is a sweet spot with subagents and parallelizable plans. It’s not about better, it’s about efficiency and picking the right model for the job. You can achieve the same results as frontier models wi
118.
by iwontberude 4mo ago
I was freaked out being stuck with OpenAI and Anthropic. I setup qwen3.6:35b-mlx on my Mac Studio M1 Ultra and was blown away really. I am no longer afraid that Anthropic or OpenAI will be able to control the market.
119.
by iwontberude 4mo ago
Just like all those drones we use on our adversaries. The next American civil war will definitely be fought with drones.
120.
by iwontberude 5mo ago
Also sorry for being rude, I had one job.

More ›