@Felix - How are you thinking about observability? Anthropic is very clear that evals are critical for agentic processes (your engineering blog just covered this last week). For my whole company to roll out access to agents for all staff, I'd need some way for staff (or IT) to be able to know (a) how reliable the systems are (i.e., evals), (b) how safe the systems are (could be audit trails), and (c) how often the access being given to agents is the right amount of access.
This has been one of the biggest bottlenecks for our company: not the capability of the agents themselves -- the tools needed to roll them out responsibly.
As an alum, you may have much more influence than you might expect. President Cauce (who's contact information is prominently displayed on the UW website) may be unaware that alumni are not pleased with this.
> 100% not at fault
When something significant goes wrong, it’s almost never due to a single cause; multiple factors and confounding elements are usually at play. We build better-performing systems when we examine various contributing factors and proximal causes.
The OP's argument is that the interface does not "clearly" indicate that one engine input is higher than the other. After reading the article and reviewing the interfaces, I have to say: the OP’s argument is quite compelling. While training is undoubtedly important and should be included in the safety recommendations, the OP makes a strong case that design improvements should also be considered as part of the safety recommendations.
I have kids who will soon be teens. They don't have phones yet, but I can see the decision looming: my ability to communicate / coordinate with them plus the fact that a huge portion of their social life will migrate online on one hand vs. all of the dangers on the other.
This is why a lot of parents start with smart watches or restricted phones. They try to get the communication / coordination benefits without the online social risks. But that only lasts so long.
I'm not sure how I'll navigate it. Probably not by saying "no to that stuff until 18".
One of the interesting ironies of philanthropy (of which there are many) is that people who have less money tend to be more generous philanthropically (as a % of adjusted gross income) than people who have more money. And because of the way the numbers work out, they give out a lot. If Upsolve can help a very very very large number of people get through bankruptcy, they might be able to amass a very strong "Pay it forward" revenue stream. Not sure if enough to fund the entire org, but enough to help out.
@rpavuluri: If at any point you were to move forward with white labeling and selling that, I would recommend you consult a lawyer about whether the IRS would consider this unrelated business taxable income (UBTI). Too much UBTI can put your 501c3 status at risk. IANAL / IANYL / YMMV / etc.
I challenge you to consider whether your perspective applies to the 99% of people who use Slack that are not developers writing software. For those people, email is often not a place where well-argued discussions happen. It's usually a place where chat-like communication happens, slowly, and in an ill-fitting interface. For non-developers, let's take a operations associate for example or a logistics manager, or someone else whose job is to coordinate and communicate frequently, Slack plays a very different role than it does for devs.
Unfortunately for you, and others like you who don't like Slack, if 99% of your company wants to use it, you'll almost certainly get pulled in. Unless your department or team explicitly opts-out, you'll get pulled into the same platform the rest of the company is using.
This isn't to say you're wrong -- just to say that we must not judge the hype around Slack, or it's clear successes, as a fad if we're only judging it based on the experiences of the kinds of people who frequent HN.
Make sure you are only including positions that help you. Common mistake I see is including tiny 2-3 month long positions, without considering whether including them is additive or not. It's often not.
short sighted, but that's sometimes the right way to prioritize. sounds like kenning is in prototyping mode, and could be intentionally prioritizing time to first test, acknowledging that s/he might have to rebuild things later. in the early stages of a startup, speed to insight can easily be more important than long term coding efficiency. so short sighted thinking can make sense.
Looks awesome. I'd love to give it a try with my team. Would you be open to extending the free teams plan to high tech nonprofits in addition to the access you provide to students, academics, and open source projects? We tend to work on distributed projects frequently with industry experts doing pro bono work for us, and something like comet could simplify our collaboration. The size of these pro bono project teams tend to ebb and flow, much like an open source project, so effective collaboration tools are critical for ramping up new folks as well as retaining learning when folks cycle off.
I'm not sure if this is your intention or not, but your comment reads as if to indicate that you believe so strongly that markets are efficient that you are unwilling to consider data that suggests an inefficiency exists.
How do the economics work out for the clients? let's say we're just doing landing pages all month long (for example). If they turnaround in two days, that's about fifteen landing pages, which at $160 for the month comes out to about $11 each. Obviously this is a contrived and unrealistic example, but am I getting the math right?
I guess this is unlimited the way my data plan is unlimited: with rate capping?
"Perhaps the female intern should not have had so much to drink." < This is the exact place where you blamed the victim. Subsequently you also blamed alcohol.
Is Android inextricably coupled with the Google App store? For thought experiment's sake: could Microsoft fork Android and then host their own app store that supports the same apps that a dev would submit to Google's App store, thus making it possible for devs to get their android apps listed in both stores merely by submitting the app twice?
(Serious question. I know nothing about app stores and/or mobile operating systems)
I disagree with your decision on what feature set was necessary to be launch ready. Not to put too fine a point on it but I think it's worth saying out loud that software that users can't quit out of is not ready for launch, especially if your docs clearly state that quitting the app is required for uninstallation.
This has been one of the biggest bottlenecks for our company: not the capability of the agents themselves -- the tools needed to roll them out responsibly.