sgrove
Joined 18 years ago
Karma
4,382
Submissions
933
Member since
Mar 16
Recent submissions
Comment · 4 months ago
Likely many points along the pareto frontier.
Some will take greater risks and win (or lose); others will play it safer and slowly accumulate wins (or be obsoleted).
Never mind the threat of letting these models write code that runs your business, or operate it agentically. Models trained by actors (corporate or nationstate) diametrically opposed to your interests.
Lots to take into account now, interesting time to be in business.
Comment · 1 year ago
There's a followup study to identify the actual cause of such a surprising outcome https://www.arxiv.org/abs/2506.19823
The combined use of faithful-chain-of-thought + mechanistic interpretation of LLM output to 1.) diagnose 2.) understand the source of, and 3.) steer the behavior is fascinating.
I'm very glad these folks found such a surprising outcome early on, and it lead to a useful real-world LLM debugging exercise!
Comment · 1 year ago
I think it's not going well? I keep getting to the start a new call page, it fails, and takes me back to the live page. I assume your servers are on fire, but implementing some messaging would help ("come back later") or even better, a queueing system ("you're N in line") would help a lot.
Really looking forward to trying this out!
Comment · 1 year ago
Or even run doom in TypeScript's type system!
Comment · 1 year ago
All of your examples have had multiple cases of going down, some for multiple days (2011 AWS was the first really long one I think) - or potentially worse, just deleting all customer data permanently and irretrievably.
Meaning empirically, downtime seems to be tolerated by their customers up to some point?
Comment · 2 years ago
I asked ChatGPT to rephrase your emotive reply into an expression of personal preference (which I think you were trying to express):
> I feel that transforming such a wonderfully written story into a podcast with robot voices might result in a shallow and awkward rendition that doesn’t capture the essence of the original work.
I can definitely agree with that in the general sense! But I have a long commute, and I thought the content of it would still come through well enough, and I've appreciated it for a few other subjects. It's not as good as having professional voice actors read it out, but it still has some value to folks!
Comment · 2 years ago
Does anyone have a pdf of all of the pages together? This seems great to listen to as a NotebookLM podcast on a commute to work on Monday.
Better yet, if someone has already _done_ that Notebook Lm podcast and has a link to it, please share!
Comment · 2 years ago
Interesting! Always interesting to see the ideas in the air at the same time!
Definitely think this sort of idea could become the "serverless" equivalent for ml-using apps. I'm curious what you think re: versioning, consumption from various client languages, observability/monitoring/queueing, etc.? Feels like it could grow into a meaningful platform.
Comment · 3 years ago
My absolute favorite overview of this - the video goes over both the explanation of the spaces, but also how it relates to a sort of understanding of reality itself!
Comment · 3 years ago
That sounds pretty fun (once you're finished with it, I'm sure)! How does sniffing the USB work? Do you do that via some software/kernel extension, via special hardware, or something simpler? Do you find there are some USB devices where the manufacturer would rather you didn't sniff their traffic and make it more painful to piece together?
Comment · 3 years ago
Yes, that wasn't a priority here, but I also don't think it's much of a concern with e.g. GPT-4's `system` vs `assistant` vs `user` roles. Would be another thing to work on, but nothing worth doom and gloom.
Although, 'script(/injection) kiddie' will be an interesting phenomenon in the future...
Comment · 3 years ago
I’ve been doing the same thing with a number of projects, building chains of prompts from one api call to another e.g. for ConjureUI (self-creating, iterable UIs that come into existence, get used, then disappear) https://youtu.be/xgi1YX6HQBw how it works to generate a full self-contained react component:
1. Take user task
2. Pass it to a prompt that requests a Product UI description of a component
3. Pass 1+2 to another that asks for which npm packages to use
4. Pass 1+2+3 to a templated prompt to write the code in a constrained manner
5. Run 4 in a sandbox to see if there are errors, if so pass it back to #4, looping
It’s currently quite slow, but that’s an implementation detail I think.
Thinking at https://linzumi.com
P26 YC. Reach me at sean@the thinking site above.
My hn-users must-follow list:
They're from the future:
* richhickey https://news.ycombinator.com/threads?id=richhickey
* brandonbloom https://news.ycombinator.com/threads?id=brandonbloom
They understand the future and create it:
* swannodette https://news.ycombinator.com/user?id=swannodette
They're bringing the future to the masses with infinite patience:
* chenglou https://news.ycombinator.com/threads?id=chenglou
* sbensu https://news.ycombinator.com/user?id=sbensu
* stopachka https://news.ycombinator.com/user?id=stopachka
* tlrobinson https://news.ycombinator.com/threads?id=tlrobinson
They're just generally interesting:
* stcredzero http://news.ycombinator.com/threads?id=stcredzero
[ my public key: https://keybase.io/sgrove; my proof: https://keybase.io/sgrove/sigs/wb_3GBhKpBugcFtZXC7MC8D0o7LHG1rO90aLrLDII3g ]