It's all really happening. Pretty much every meeting with friends touches on what their recent sightings or stories of ICE terror have been. Everyone who hasn't seen it first hand has a second hand story.
Absolutely everyone I talk to is against ICE's actions and that is the thing giving me hope that it will be defeated by the citizenry.
Essentially, you're asking the LLM to do research and categorize/evaluate that research instead of just giving you an answer. The "work" of accessing, summarizing, and valuing the research yields a more accurate result.
Thank you so much for sharing this. Myself, and I’m sure many of others, are thinking about these things a lot these days. It’s great to see how someone else is coming at the problem.
I love the grounding back to ~“well even a human would be bad at this if they did it the current LLM way.”
Bringing things back to ground truth human processes is something that is surprisingly unnatural for me to do. And I know better, and I preach doing this, and I still have a hard time doing it.
I know far better, but apparently it is still hard for me to internalize that LLMs are not magic.
In exploring `gh` CLI extensions recently I found at least 2 devoted to being a better PR review interface. I've also seen [Trag](https://usetrag.com/) and others pop up.
But then this analogy doesn’t hold… Has there ever been a startup in history where it’s break even everywhere? Investors don’t get their money back, customers (hopefully!) are made whole, and the founders and employees are now out of a job and _perhaps_ didn’t get their final pay check depending on how bad it is.
Yeah, you're missing my point. Given most startups fail, the question isn't "at the end of the day do you still get a paycheck for your failed startup", it's "how much of your life did you burn on that failed startup".
Since we're talking about a specific startup whose founders are participants here, I think we can do without the ghoulish stuff about them not making payroll or whatever; "winding down" implies they're failing in an orderly way.
I got your point (after the Free Parking clarification).
(To be clear, my comment wasn’t on Double specifically. No clue how strapped for cash they are while winding down. They seem to be doing right be people (paying fees and such), and that’s great.)
Finally, maybe it’s unintentional, but you seem to be implying that “it’s not worth burning your life on a failed startup,” which seems like a bad take.
If you spend 5 years on a startup that shows promise but ultimately doesn’t pan out, is that always worse than spending 6 months on a startup that fails fast? First, this would be wildly hard to prove, and second, there are obviously counter examples.
In that 5 years, I’ve identified a shit ton of stuff that works, doesn’t work, and made relationships along the way. I’d bet money on the founder of the 5 yr startup being successful over the 6 months founder.
You do that stuff in every job. Meanwhile, people with careers in startups do multiple startups. You get a finite number of them. You are literally better off working at somebody else's successful startup than you are spending 5 years on a doomed startup. The most valuable asset you have is time.
I really don't think I need to cite a study that 5 years of time is a lot to give up on a company that goes nowhere. It happens! It's normal! And very painful. Which is why getting to a decisive resolution in 6 months is often a gift.
You will make more money, and probably work on equivalently interesting problems, working any other job than at a startup you cofounded that limps for years before winding down. It's really hard to see what upside you're finding here. I don't say this often, but this isn't a place I see leaving at "reasonable disagreement".
> I really don't think I need to cite a study that 5 years of time is a lot to give up on a company that goes nowhere.
I think your problem is that you have one definition of success for a startup and that’s to become a unicorn.
> working any other job than at a startup you cofounded that limps for years before winding down.
Not everyone starts a startup to become a billionaire, and not everyone seeks to _make more money_.
But aside from that, the way you’re going to _make more money_ in another job is by working at a large company that can pay you more money. That large company has lots of people and lots of jobs that are specialized… because? Lots of people are there to do them. If I work on a team in a large company, I very likely have a narrow focus and I am not “solving interesting problems” on average.
You keep saying stuff like this, but I'm asking a simple question. How have you personally benefited from spending years at a startup that failed, so much so that you were better off at the doomed startup than in any other place in the industry? I'm looking for some kind of relatable experience.
I’ve worked at a few “unsuccessful,” but long term startups. Bought, but kind of for parts. The 2 yo startup I closed down was the most enjoyable and most rewarding from a learning perspective. The longer term, but unsuccessful, ones were also enjoyable and great learning experiences.
The two larger company experiences had me much more pigeonholed, even as a principal engineer / architect, and exposed me to far less of the important aspects of running a company or product, generally, but I made money. Often it was not enjoyable, and certainly leadership did not always enjoy my questioning of their actions and decisions.
During what "era" did these startups wind down? Pre- or post- first Internet bubble? How old were you at the time? Maybe a failed startup is a great experience when you're 23, and not so much when you're 38?
(I've worked in startups my entire career, and been a founder of five, one serious acquisition, two going concerns, and two failures, one fast one slow; the long failure, circa 2001, was not a valuable experience in any way.)
I started working at startups in 2009… in my late 20s.
If you’re not already financially stable at 60 and hoping to win the startup lottery to retire… well…
So yeah, obviously age plays some role here.
I didn’t go work at startups because I thought I was gonna get rich. I worked at startups because I expected to learn waaaaay more than at Big Co, and have wayyyy more fun while doing it. This panned out.
Money has never been my motivation in computing—interesting problems are.
It wasn’t until I turned 42 that I realized I did it all backwards, though. I should have started by working at Big Co and saved/invested every penny to seek early retirement, and then do whatever I wanted with no financial pressure, later.
I could have stayed at a Big Co for 4+ more years and racked up more RSUs, and bonuses, but my job became boring, and I started to resent it. What can I say?
Curious about the 2001 failure… but am guessing the amount of free money in the dotcom boom led to you trying to build something not well thought out or with ang real hope for viability. (not throwing shade, but lots of mud was thrown and very little stuck)
I don't understand this "startup lottery" stuff. I've been on HN since roughly 2007 and for most of those 18 years I've been a bootstrapping evangelist. My most successful exit to date was a bootstrapped company. I'm just saying: there's no romance or great reward to working in a failing startup. If you have a clear signal that your startup is failing, you've been given a gift from the heavens: the ability to see into the future, see yourself wasting years, and then navigate around those years.
2001 was the beginning of the "nuclear winter" for startups. I think part of this is that I'm just older than you? (I have no idea how old you are, but "how did your startup fail in 2001" is an odd question).
Were you a founder at any of these startups you're talking about? If you're just talking about being an employee, that's totally different. It's often (maybe even usually) reasonable for employees to value their equity at ε. That being the case: you can absolutely choose roles based solely on base comp, how interesting the work is, and how it looks on your resume.
Finally: if you try to factor money out of entrepreneurship discussions, you get to very funny places. We're in one of them now! It is strictly better for a startup to fail at 6 months in than it is for it to fail at 5 years in.
When you work with Apex (or really any other technology bridge to the traditional financial world, Q2's Helix is common for traditional banking, Apex Clearing is common for stock trading, etc) they require you think about things like this during your implementation. Its not quite as turnkey as something like opening a Stripe account; your implementation will need to demonstrably pass a playbook of tests before your partner will allow you to play in real financial transactions - and those tests typically include things like account closure or program shutdown.
Basically, the traditional financial services partners who give startups access to these legacy networks know their clients are startups who might not fully understand the space or might want to cut corners. They're good at making sure they're protected against their clients' behavior, and in most cases legally the end users are actually the customer of the financial services company, the startup will be considered a "deposit broker" instead of a "bank" etc. Its been longer since I've touched the stock broker side so I'm fuzzy on the specific terminology but its similar there.
Reminds me of [Language as Intermediate Representation](https://chrisvoncsefalvay.com/posts/lair/) - LLMs are optimized for language, so translate an image into language and they'll do better at modeling it.
Cool connection, hadn't seen this before but feels intuitively correct! I also formulate similar (but a bit more out-there) philosophical thoughts on word-meaning as being described by the topological structure of its corresponding images in embedding space, in Section 5.3 of my undergrad thesis [1].
I think an alternative is to fund _individuals' maintenance of the projects_, as opposed to the project itself. Filippo Valsorda has written about this recently: https://words.filippo.io/full-time-maintainer
That's an interesting idea, thank you, I'll read through the post.
EDIT: I found the post interesting but unfortunately these projects are in no way load-bearing. I'm happy to hear Filippo has made it work for himself but as he points out, the projects he works on are simply much better suited to sponsorship and retainer/consulting agreements. I wonder what he would recommend for newer or less-load-bearing projects.
> Let's make a timeline chart of https://www.githubstatus.com/history for the past 1yr and upload it as a gist
yields [GitHub Status Incident Timeline — Feb 2025 to Feb 2026](https://htmlpreview.github.io/?https://gist.githubuserconten...)
219 total incidents across 12 full months, averaging 18.3/month. January 2026 was the worst month, and August 2025 was the calmest.