ryoshu

5 ms·

31.
by ryoshu 9mo ago
To pedant it up, not across GPUs.
32.
by ryoshu 10mo ago
Tulsa used to have a rich Black neighborhood.
33.
by ryoshu 10mo ago
SAP or Salesforce can do anything. Or you can write custom code for a fraction of the price for small/medium sized businesses.
34.
by ryoshu 10mo ago
If he's having an impact it's because he can break through the bureaucracy. He's not trying to protect a fiefdom.
35.
by ryoshu 10mo ago
LLMs fail at causal accuracy. It's a fundamental problem with how they work.
36.
by ryoshu 10mo ago
I can't do that either without opening up an image editing tool. Give the model a tool and goal with "vision". Should work better.
37.
by ryoshu 10mo ago
The reward is helping people hone their craft and grow in their careers. Being a shit shield sucks, but the trade-off is worth it.
38.
by ryoshu 11mo ago
Some people write to convince. Some people write to provoke.
39.
by ryoshu 11mo ago
They should audit LLMs.
40.
by ryoshu 1y ago
Agreed. I use CC a lot for exploratory work. It's great with fast iteration for throwaway code.
41.
by ryoshu 1y ago
You can look it up on his blog: https://www.antipope.org/charlie/blog-static/2006/06/
42.
by ryoshu 1y ago
This article talks about prompt evals https://www.anthropic.com/engineering/writing-tools-for-agen... . There are plenty of approaches to provide some degree of rigor around the slot machine output.
43.
by ryoshu 1y ago
Data vs. feels
44.
by ryoshu 1y ago
The Ren Faire in NY has a time traveller day. Trekies abound.
45.
by ryoshu 1y ago
I asked GPT-5 to generate an image prompt. I asked it to use that prompt to generate an image. It was a content policy violation.
46.
by ryoshu 1y ago
And you have to pay special attention to the tests written by LLMs. I catch them mocking when they shouldn't, passing tests that don't actually pass, etc.
47.
by ryoshu 1y ago
A lot of times coding is thinking. It's like sketching out an idea on paper, but with code.
48.
by ryoshu 1y ago
Used to be a mistake. Now it's pretty clear you have to pay-to-play to get top billing on SERPs.
49.
by ryoshu 1y ago
Last attribution doesn't work. Retargeting is a joke and the most valuable spenders have ad blockers so you can't even reach them through display ads. The data is manipulated everywhere as long as you can hand wave towards ROAS.
50.
by ryoshu 1y ago
Possibly https://www.secretservice.gov/investigations/cyber
51.
by ryoshu 1y ago
Optimize everything so you can take the soul out of it.
52.
by ryoshu 1y ago
The demo videos show very poor knife skills.
53.
by ryoshu 1y ago
Reverse centaur.
54.
by ryoshu 1y ago
I still struggle with ORMs. SQL is... declarative. If you're working with multiple RDBMSs, sure? Maybe I want my local dev to be sqlite and scaled be postgres? I've never run into that in production. A DSL on top of a DSL doesn&#x
55.
by ryoshu 1y ago
"qual"
56.
by ryoshu 1y ago
I visited Dachau years and years ago. It was a nice summer day, but a pallor fell over when we went inside the camp. It felt like the sky darkened and the color drained from the entire environment.
57.
by ryoshu 1y ago
Selection bias + perceptual adaptation is my experience. Selection bias happens when we play the probabilities of using an LLM and we only focus on the things it does really well, because it can be really amazing. When you use a model a lot
58.
by ryoshu 1y ago
They do. I watched a person quit because they had to hand edit video frames that came out of AI. It would have been cheaper to do traditional VFX.
59.
by ryoshu 1y ago
Pretty dang close isn't the same as accurate for an exchange of time and money. Voice->text, with a noisy background, is a particularly hard problem. Especially with hardware not designed to limit background noise. Try it. Whisper i
60.
by ryoshu 1y ago
Social media and phones with cameras has made a lot of people risk adverse. When a video can hit Twitter or TikTok or Reddit about something that can ruin your life why risk it?

More ›