verdverm

9 ms·

181.
by verdverm 26d ago
More people can serve up static markdown for the agents with `Content-Type: application/markdown`, CDNs work really well. I think both sides of this dynamic would find it beneficial. I suspect much of the infra stress is on all the &qu
182.
by verdverm 26d ago
With opencode, I just start a new session and it's the same blank slate, there's no lingering of prior state Does /clear in claude not do what I intuit it to?
183.
by verdverm 26d ago
100% I am big believer (based on personal experience) that there is a ton of ROI in harness engineering, to go along with the context engineering, invariably intertwined
184.
by verdverm 26d ago
> just tediously inconvenient Given all they claim about their Ai's capabilities, they ought not be able to use this as an excuse
185.
by verdverm 26d ago
I didn't want to use "gatekeeper" because I'm biased and that's how I see Big Ai and what I perceive to be attempted regulatory capture Suggestions for an alternative?
186.
by verdverm 26d ago
What about phone cameras? I'm curious if/how people make a distinction, I really like the UX of AR (as found in the HoloLens 2), hardware is getting close to where I might pick something up again (FOV has hardly budged, form facto
187.
by verdverm 26d ago
I think you are underestimating the diversity of systems and tools developers use > who all use codex I do not > We all have xcode I do not
188.
by verdverm 26d ago
costs are in a large part tied to model parameter count, eg. a small model fits on one GPU, the biggest models require an entire rack the primary difference you see through those prices is model size that you don't seem much capability
189.
by verdverm 26d ago
Import bans, like we have on Chinese EVs in the US. The US already restricts other countries access to Huawei hardware This is another example where we might ask "why would it be any different?"
190.
by verdverm 26d ago
Bitbucket Cloud (Atlassian)
191.
by verdverm 26d ago
I suspect that the growth in LLM "crawlers" is part traditional crawlers and a growing part WebFetch tools agents use directly on behalf of users. Personally, my page fetch count is way up because I have a custom deep research age
192.
by verdverm 26d ago
<moved to description box>
193.
by verdverm 26d ago
I was even thinking this while spelunking the OpenCode repo, there is a ton of complexity in there for things I care zero about, like powershell support, which created a situation where they dynamically construct the tool description and I
194.
Ask HN: HIPAA for AI vs. Licensed Providers, your thoughts?
1 points by verdverm 26d ago|7 comments
195.
by verdverm 26d ago
one can just not add the ai to the call, yea? the major call providers have settings one can use as well
196.
by verdverm 26d ago
just ignore grandpa, more and more people are
197.
by verdverm 26d ago
Instructing to use the `gh` CLI, not ideal for those of use using BBC at work They also have instructions about how to format certain output, which conflicts with the instructions we have in repo. I only discovered yesterday because we were
198.
by verdverm 26d ago
this is why open models are important, you should be able to adjust the model's tone, without having to go through a gate-keeper, where they need to have a "societal" level alignment (impossible broadly, applicable to certain
199.
by verdverm 26d ago
and thus far, the courts have disagreed, Ai output is not copyrightable, and (I'm hypothesizing) not protected speech either
200.
by verdverm 26d ago
we are typically not in the position to judge our own tropes, like a racist doesn't think they are racist, it's very hard to recognize our own biases and gaps you are missing important facts and say the things about China that are
201.
by verdverm 26d ago
personalized software, for people and companies, i.e. the SaaSpocolypse still not clear how this will play out, I have more personal software, but I was doing that before LLMs were an everyday thing
202.
by verdverm 26d ago
turns out LLMs do not make everyone an expert, it takes a real expert in a field to get value out of them see Terence Tao's convo with ChatGPT, I certainly could not have that kind of a session around mathematics
203.
by verdverm 26d ago
I suspect they are trying to reorder regulatory capture and wall street debut
204.
by verdverm 26d ago
every model does this, it's their jam
205.
by verdverm 26d ago
> Fundamentally I don’t think we have a choice about riding on the AI technology train. It’s a wild ride and I just hope we’ll get through it OK. > I don’t like them. They talk to me in this grating LLM-voice, an uncanny valley of tal
206.
by verdverm 26d ago
Has Zenodo become the place where "citizen scientists" post their Ai generated papers since Arxiv put the vouching in place? I've seen more HN post from this site to questionable papers
207.
by verdverm 26d ago
the grift must grow?
208.
by verdverm 26d ago
look into the prompts, there are some choices in there disclaimer, I use opencode and have customized parts of it, and will do more, but it is a solid foundation and comes with more out of the box than pi pi is too minimal for me, I'd
209.
by verdverm 26d ago
Coming Soon: https://artificialanalysis.ai/agents/coding-agents?coding-ag... They used to have one chart with {model X harness} for a subset of combos, looks like that is getting an upgrade
210.
by verdverm 26d ago
so many tropes man... getting a non-american perspective on what's happening (for just about anything) is a good thing to do

More ›