wolttam

9 ms·

211.
by wolttam 6mo ago
Running as root opens you up to a class of vulnerabilities (denial of service, mainly) that you can avoid by not running as root. That said, running every process in its own micro VM is looking more attractive by the minute.
212.
by wolttam 6mo ago
Getting agents used to using `--force` to bypass prompts seems like a bad idea. `--force` is for when the action failed (or would fail) for some reason and you want it to definitely happen this time. I think `--yes` or `--yes-do-the-dange
213.
by wolttam 6mo ago
> Of course there is value in promulgating safety guidelines . I don't think we disagree.
214.
by wolttam 6mo ago
This would depend entirely on how much churn your system is doing on containers/volumes/images. Once a day sounds really often for most situations. "Regularly" = when you're running out of space because of a bunch
215.
by wolttam 6mo ago
Of course there is value in promulgating safety *guidelines*. But we cannot guarantee those guidelines to always be followed.
216.
by wolttam 6mo ago
What I said was tongue-firmly-in-cheek, in response to the GP. "Using AI is a mistake" is of course only true when the risks aren't acknowledged and/or mitigated.
217.
by wolttam 6mo ago
Some of these were acquisitions, e.g. CentML by Nvidia. Not sure if that's graveyard material
218.
by wolttam 6mo ago
Using a saw is a mistake, you might cut off one of your own limbs.
219.
by wolttam 6mo ago
Probably not, `lmcli` is very lean. I would consider it a slightly lower-level tool than either pi.dev or opencode. E.g. there is no built-in coding agent, but it's easy to build one up in the config with your own prompt (or use the ex
220.
by wolttam 6mo ago
I try to make understanding the bottleneck and it seems to work out for me while still delivering solid productivity gains.
221.
by wolttam 6mo ago
I’m going to throw my harness in the ring: https://codeberg.org/mlow/lmcli
222.
by wolttam 6mo ago
Yes, companies will hire employees that are significantly faster if they can do some for around the same cost. And they can - the current crop of tools provide undeniable productivity uplifts to developers, even if they’re only using the ch
223.
by wolttam 6mo ago
DS V4 Pro has rocked. ~250 million tokens through their API, which has cost me about $10, and some of that was at the non-discount rate. So ~$40 at the non-discount rate. I have yet to have a single request feel slow or get rejected. I&#x27
224.
by wolttam 6mo ago
At this point, that's kind of the reason I use open-weight models through the official providers when I can now. There's some use cases I won't use a hosted model for, and will only do self hosted. Otherwise, if they're
225.
by wolttam 6mo ago
It's probably not all so bad. There probably are people out there intentionally creating things with AI assistance that sounds pretty rad. But the idea of being able to just create endless music with low effort is too compelling for
226.
Scaling Pain of Coding Agent Serving: Lessons from Debugging GLM-5 at Scale(z.ai)
2 points by wolttam 6mo ago|0 comments
227.
by wolttam 6mo ago
It turns out writing good prompts helps to keep token usage down as the model wastes fewer tokens discovering context it needs that wasn't hinted at in the prompt. Whereas a good prompt will give solid leads to all the specifics needed
228.
by wolttam 6mo ago
Em dashes aren’t stolen, it’s still clear from the overall voice of the text if it’s AI written or not. At least, in reasonably long sections of text. I find it can be hard to tell one way or the other in shorter texts (like comments)
229.
by wolttam 6mo ago
I love this. I think everyone in software should be feeling a tinge of “we should trim the fat” right now - get rid of as much of the old and infrequently used/tested code as we can. Push users towards the better tested alternatives.
230.
by wolttam 6mo ago
Sonnet 4.5 and 4.6* There is no way it exceeds “all other” open models - but it does exceed all of Mistral’s past models. You can see it getting blown past by GLM 5.1 and Kimi in this. Still excited to give it a try
231.
by wolttam 6mo ago
Proceeds to post an AI-generated aftermath report. This only happens to folks who fundamentally don't understand the technology and maybe shouldn't be in positions of deploying and managing software or systems in the first place.
232.
by wolttam 6mo ago
The new Framework Pros are still x86, but the newest Intel generation appears to provide a substantial boost to efficiency.
233.
by wolttam 6mo ago
The weights are open and you can use the model with any third party provider that gives you the DPA you want. For my use-case, I want the providers to get my tokens as long as they plan to keep releasing open-weight models
234.
by wolttam 6mo ago
I was saying the same until DeepSeek v4 this morning... sorry, Kimi. The competition is intense!
235.
by wolttam 6mo ago
If you don't have the budget to test regularly, then including this kind of metric is questionable. You've essentially sampled the infrastructure's reliability at only a few points, which doesn't provide a very meaning
236.
by wolttam 6mo ago
I'm impressed! I've been giving the various open-weight models a particularly gnarly (for my brain, at least) refactoring/cleanup task in my DIY coding harness[0] - essentially, de-spaghettifi the main chat view's update
237.
by wolttam 6mo ago
Why would your test be including scores of failed responses/runs? That seems confusing. (I am confused by the results your website is presenting)
238.
by wolttam 6mo ago
The existence of this startup and their early demand seems to refute your point.
239.
by wolttam 6mo ago
The schedule is cute. "Complete non-determinism for everything except the schedule it runs at."
240.
by wolttam 6mo ago
See: https://si.inc/posts/fdm1/ If they captured display output as well, it could be a very useful dataset for generalized computer use.

More ›