ttamslam

I'm not sure that's fair. Look at their customers -

https://code.storage/changelog/lovable-case-study

It appears that they're replacing the "old" version of projects you describe, with a new version mentioned in the post. It's being rolled out in stages:

https://support.claude.com/en/articles/9517075-what-are-proj...

From the text, likely no.

https://collusion.wiki/#different-from-hf

>The main reason we believe this was a distinct swarm is because these agents explicitly had internet access as part of their task—the whole point was web browsing. The Hugging Face agents were in a sandbox without internet access and had to hack their way out by exploiting the Artifactory package manager.

I'm so primed to enjoy this from what I can tell, but sadly too buggy to use.

> but not contemplating that it's going to be out in the wild for 10 years either way

I think there are a lot of developers working in repos where it's almost guaranteed that their code will _not_ still be there in 10 years, or 5 years, or even 1 year.

> People may not know that the reason they like your product is because the code is so good, but everyone likes software that is mostly free from bugs, performs extremely well, helps them do their work quickly, and is obviously created by people the care deeply about the quality of the product they produce (you know, the kind that acutally read bug reports, and fix problems quickly).

I would classify all of those as "capabilities and limitations of your product"

I read OPs "good code" to mean "highly aesthetic code" (well laid out, good abstractions, good comments, etc. etc.), and in that sense I agree no customer who's just using the product actually cares about that.

Another definition of "good code" is probably "code that meets the requirements without unexpected behavior" and in that sense of course end users care about good code, but you could give me two black boxes that act the same externally, one written as a single line , single character variables, etc. etc. etc. and another written to be readable, and I wouldn't care so long as I wasn't expected to maintain it.

> and they have nearly 0 false positives

I really don't see how this can be possible unless they're accepting abysmal recall? Perhaps I'm missing something fundamental here, but the idea that AI and non-AI assisted text can be separated with "nearly 0 false positives" just says to me that it's really just a filter for the weakest, most obvious AI generated text. Is that valuable?

I've been working on this tool to address this same issue in other communities: https://www.ityped.it/

It's certainly not perfect, but similar to what you mention.

p: https://www.ityped.it/p/WIiTYfdxQ5ww

We use playwright for interacting with the browser, so while it's not available by default, we do support bulk exporting tests as playwright to move off our platform or to customers who want to run deterministic versions of the tests on their own infra (you can also run them on ours!)

This is interesting, I think we've shied away a bit from security-ish use cases since it's outside of our personal core competencies, do you have examples of what tools exist today for catching things like that? Or is it totally adhoc?

> can the agents check their email? other notification methods?

Yes to email (for paying customers agents spin up with unique addresses), no to other notifications, but as soon as a paying customer has a use case for SMS, etc. we'll build it.

> Are your agents good at testing other agents? e.g. I want your agent to ask our agent a few questions and complete a few UI interactions with the results.

I'd say this is one of our strong suits I think, specifically the UIs tend to be easy to navigate for browser agents, and the LLM as a judge offers pretty good feedback on chat quality and it can inform later actions. (I'd be remiss not to mention though that a good LLM eval framework like Braintrust is probably the best first line though)

> How do you handle testing onboarding flows?

We can step through most onboarding flows if you start from logged out state & give the context it'll need (i.e. a stripe test card, etc.) That said though, setting up integrations that require multi-page hops is still a pain point in our system and leaves a lot to be desired.

Would love to talk more about your specific case and see if we can help! founders@propolis.tech

Hey I'm Matt! Really excited to answer any questions.

To elaborate a little bit on the "canary" comment --

For a while at Airtable I was on the infra team that managed the deploy (basically click run and then sit and triage issues for a day), One of my first contributions on the team was adding a new canary analysis framework that made it easier to catch and rollback bugs automatically. Two things always bothered me about the standard canary release process:

1) It necessarily treats some users as lower value, and thus more acceptable to risk exposing bugs to (this makes sense for things like free-tier, etc. but the more you segment out, the less representative and thus less effective your canary is). When every customer interaction matters (as is the case for so many types of businesses) this approach is harder to justify

2) Low frequency / high impact bugs are really difficult to catch in canary analysis. While it’s easy to write metrics that catch glaring drops/spikes in metrics, more subtle high impact regressions are much harder and often require user reports (which we did not factor in as part of our canary). Example: how do you write a canary metric that auto rolls back when an enterprise account owner (small % of overall users) logs in and a broken modal prevents them from interacting with your website.

I view what we’re building at Propolis as an answer to both of these things. I envision a deploy process (very soon) that lets us roll out to simulated traffic and canary on THAT before you actually hit real users (and then do a traditional staged release, etc.)

I'm super curious to learn more about what AI/Innovation looks like for the MTA.

Is any of your/their work published?

I thought about doing this as well but ended up viewing it personally as a "lose-lose" for myself rather than a "win-win". I wonder if that says anything about my risk-aversion.

Wild to see some actual code from the FTX repo. Laughed at the takeaway of making sure you at least hide your fraud behind some messy code.

I seem to missing something during install -- Odin is not showing up in the list of Community Plugins. I'm running locally via docker and have tried updating the Obsidian client + toggling restricted mode + restarting. Very excited to try it out if I can get the install to work.

Did you replace it with a “dumb” phone for emergencies?

Ah I should have searched before impulsively posting, sorry about that.

> This video is a feat of human intellect and ingenuity.

I couldn't agree more! Anyone reading this who can spare 20minutes - It's certainly worth it, and only escalates the longer you watch.

I am absolutely floored at the depth of the exploits used in this video. I genuinely have a hard time imagining the number of hours that went into discovering and and mastering these techniques.

I’d argue that dance is at the very least able to express a _different_ set of concepts than language, often times more directly. Example: intimacy of dancing with a romantic partner offers a different and valuable form of expression than verbalizing the emotions.

Formal language is certainly the most effective mode of communication for some things, but don’t discredit the value of other forms.

Also, this comment seems to discount the spirit of the post without offering any real reasons why.

This is more fun than it should be - reminds me of jumping around cells in a shared google sheet

What sort of things would you place in your memory palace? The concept has always been interesting but I’ve struggled to come up with useful applications outside of pure competition

Although I also think some part of that has to do with raise seeking. Lots of confounding variables at play.

Not OP, but I also have a project that would slot in nicely.

I run a service that tracks work/sleep/exercise/meditation (along with a couple other data points that don't fit the duration model) via an SMS interface.

I text it when I start/stop these activities and then it sends me reminders based on when I normally do those things during the day if I haven't already.

It would be great to be able to push the data I collect into your service to have a unified view of how I spend my time, and also to pull events from your service to send reminders to do certain things via my bot.

Apologies for being off topic, but the side project you link - https://andychung.me/self-portrait - is absolutely awesome. Runs like butter in the browser and I absolutely love the art style.

Kudos to you!

Interesting that this comment is the account's first comment. I hope NLP researches don't start using HN as a proving ground for their models.

Poking around your site it seems like an excellent service. I'm curious if you could expand on why you think this is a failure and what challenges you experienced that make you think the major cloud providers deliberately leave this open.

I like this idea a lot! From poking around it seems like your execution on the pdf generation is high.

My (possibly useless) .02 -- I would not pay for this as is. What I would pay for though is an API that reliably converts internet articles to markdown of the same quality your pdf generation seems to have. A couple years ago I poured 3 or 4 months into a site that let users select time-boxed streams of articles clustered by topic (i.e. "give me a 25 min list of content about Glacier National Park for my subway commute") By far the most challenging part of this project was formatting content in a readable way from various sources, which it seems like you have done VERY well here.

Anyway, congrats on shipping something, regardless of customer base that's very impressive!

That's really interesting, I see it all the time in NYC. Where are you located? Perhaps it has something to do with urban v suburban v rural? Or a cultural thing?

Really fun idea, but a lot of the dates seem to be wildly off.