If there was serious interest, they would taken a next step by now. They won't give you a hard no unless you're an absolute joke.
They will wait and see if there's any deal heat.
Are you talking to other funds? You need to talk to as many funds as possible in a 2-3 week period to create leverage. Do not talk to a single or small number of funds in a process[0]. Best case they will snake the round at a discount, worst case you'll give up a ton of leverage and kill the process.
[0] unless you have close relationships with a stable of funds. you do not.
I've built up to a very similar process, but it looks like yours is a much better oiled machine. Specifically I struggle with getting enough architecture thoughts in place for the programming agent to really do what it needs to do without me going behind it and refining its work. Your research repo is incredibly inspiring and has given a lot of think about.
It's sort of surreal to feel the change in the software development lifecycle over the last 24 months - what you are describing will very likely become the norm amongst most developers.
> Cursor is happily charging me something like $100 a day. If anyone from Cursor is reading this – is there a “solo dev building absurdly large systems” discount tier I’m missing?
I'm also paying a similar bill (but honestly I think it's incredible value). I'm curious about this comment though - I picture LLM pricing as consumption rather than per-seat (token in/token out) - would it really make sense to offer volume discounts on a single seat versus total volume? These platforms shouldn't really care about how many developers are consuming the tokens, just the total consumption, right?
Sounds like you’ve been bumping into the same wall and asking the same questions.
Since I switched to this layered setup, the amount of “cleaning up after the programmer” dropped a lot. The programmer‑AI has become much more obedient, and most of the real fights now happen with the architect‑AI instead.
You’re absolutely right that it’s hard to front‑load enough architectural thinking. That’s also why this project raced its way to version 12 so quickly – every time the architecture felt wrong, I forced myself to rethink and re‑document it, not just patch the code.
On the token/pricing side, I honestly haven’t thought too hard about the exact math. One way or another, we’re all just helping to pay Nvidia’s bills, right?
Agree, but for example, migration scripts are still often just a bunch of long .sql files (unless it's Liquibase with its own cross-DBMS XML syntax), or test/staging/benchmark schemas. Even today.
And subling commenters say that all you need is raw SQL and results mapping to your code. Which I did for a while, but found that mapping is a lot of copy-pasta with minor diffs, a burden to maintain. So it's easier to use a thin library like JOOQ for mapping, or use only the mapper part of a bigger ORM framework like Django/Hibernate.
And my argument is that it's easier to map to/from a concise strongly-typed ABI/API structs instead of one raw SQL string with its structure designed for human reading/writing, like SELECT before FROM. There are such ABI-s, but they are DBMS-specific, while SQL is less so.
You may not need to use an ORM, but hand writing SQL, especially CRUD, should be a terminable offense. You _cannot_ write it better than a process that generates it.
> Stripping information from an identifier disconnects a piece of data from the real world which means we no longer can match them. But such connection is the sole purpose of keeping the data in the first place.
The surrogate key's purpose isn't to directly store the natural key's information, rather, it's to provide an index to it.
> The solution is not to come up with yet another artificial identifier but to come up with better means of identification taking into account the fact that things change.
There isn't 'another' - there's just one. The surrogate key. The other pieces of information you're describing are not the means of indexing the data. They are the pieces of data you wish to retrieve.
Any piece of information that can be used to retrieve something using this index has to be available "outside" your database - ie. to issue a query "give me piece of information identified by X" you have to know X first. If X is only available in your index then you must have another index to retrieve X based on some externally available piece of information Y. And then X becomes useless as an identifier - it just adds a level of indirection that does not solve any information retrieval problem.
That's my whole point: either X becomes a "real world artifact" or it is useless as identifier.
That's not really how data is requested. Most of these identifiers are foreign keys - they exist in a larger object graph. Most systems of records are too large for people to associate surrogate keys to anything meaningful - they can easily have hundreds of billions of records.
Rather, users traverse that through that object graph, narrowing a range of keys of interest.
This hacker news article was given a surrogate key, 46272487. From that, you can determine what it links to, the name/date/author of the submission, comments, etc.
46272487 means absolutely nothing to anybody involved. But if you wanted to see submissions from user pil0u, or submissions submissions on 2025-12-15, or submissions pertaining to UUID, 46272487 would in that in that result set. Once 46272487 joins out to all of its other tables, you can populate a list that includes their user name, title, domain, etc.
Do not encode identifying information in unique identifiers! The entire world of software is built on surrogate keys and they work wonderfully.
> This hacker news article was given a surrogate key, 46272487. From that, you can determine what it links to, the name/date/author of the submission, comments, etc.
> Do not encode identifying information in unique identifiers! The entire world of software is built on surrogate keys and they work wonderfully.
The amount of manual work required to manage duplicates is in no small part the result of not thinking enough about the identifiers and simply slapping surrogate keys on the data.
Has anybody written about why this is? I recently canceled my ChatGPT subscription that I had held since November of 2022.
I don't follow the benchmarks closely, but I know quality and speed when I see it. ChatGPT feels like its from another generation compared to the others.
Why do the stores have the coin deposit if leaving the shopping cart, even if you circumvented the deposit, is morally more morally reprehensible than urinating in public?
> Is this a US phenomenon
Yeah, you can kind of do whatever you want here. It's sort of our thing
For the uninitiated, what's the subtext here? Is this about a trend in behavior of teens in Germany?
Asking for the same reason I asked OP my question. I was just in Berlin and, second only to Tokyo, I've never* seen such conscientious group of people, including teens.
* I'd actually place two other cities from that trip into the same #2 position: Zurich and Vienna.
Yes, it's unfortunately a trend to throw e-scooters into rivers. Unfortunately the government is still not collecting stats about this vandalism. Throwing e-scooters with batteries into rivers is dangerous, because the batteries contain poisonous metals. Unfortunately many people appreciate it when people throw these scooters into the river, because they are "annoyed be them when they are standing in their way on the sidewalk".
> Promoting cart return might be as simple as setting a new norm
Nobody who does this is not aware that returning the shopping card is the normal, expected, and 'right' behavior. There just isn't a moral hazard that prevents them from doing the wrong behavior.
I think there is a larger philosophical/moral question of WHY should someone do the 'right' thing in the absense of a moral hazard. It's something I've thought a lot about over the last couple of years.
As the 4chan Shopping car theory points out, the cost of leaving the cart is zero. And the benefit is the saved time/energy. Why shouldn't a rational self-optimizing person leave the cart there? Why shouldn't they hold the subway door open to catch the train? Why shouldn't they pull up the very front of the offramp and merge at the last second? Zero cost, all benefit.
I have a self-motto of 'do the right thing' in virtually anything I do. In those examples, I'd return the carts, wouldn't hold a train door open, and would miss an exit and turn around.
But WHY do I do it? Why do I feel like I HAVE to do it? Am I actually experiencing any benefit in life over those who don't?
> But WHY do I do it? Why do I feel like I HAVE to do it? Am I actually experiencing any benefit in life over those who don't?
If I try to dig in deeper for why I also feel that way, I guess it's not about coercion or fear of judgement/retribution. I just have an innate understanding that other people have their own lives, and I don't feel like it's worth it to do things that have a minuscule "benefit" for me while being a far outweighed drawback for multiple strangers. Even though it doesn't benefit me, it does benefit the community I'm in, and is one of many things that make the society I live in relatively nice.
Not returning the shopping cart saves a rounding error's worth of time, but now multiple car drivers are annoyed in a major way when shopping carts are rolling back and forth, ramming into parked cars or taking up empty parking spots. Employees now have to spend more of their time getting all the carts, sometimes in bad weather. Not worth it.
Holding the subway door saves several minutes for me, but makes the schedule tighter for the operator and forces hundreds of people to wait a few more seconds for me. This difference between my benefit and others' drawback isn't as drastic as the shopping carts, so the bar for me to do it is lower (I would probably do it if trains were >10 minutes apart). But it also has a sketchy feeling to it - I'd trust that the train will remain stopped, but the chance of you getting caught on the side of a moving train is >0%. It has happened many times before, especially in older systems.
I don't see what the benefit is for leaving a highway at the last possible second. If anything, this erratic behavior is unexpected and is more likely to lead to an accident. Not worth it, even discounting any feelings you have for other people.
> I don't see what the benefit is for leaving a highway at the last possible second. If anything, this erratic behavior is unexpected and is more likely to lead to an accident. Not worth it, even discounting any feelings you have for other people.
In large metro areas, exit lanes can be back up, usually because there is a light at the end of the exit. For instance, exit 32 on the BQE can backup to the point that you sit in the exit lane for 10+ minutes as batches of cars move through the intersection. To circumvent the wait, some people just pull up to the front of the exit lane and merge in and go through the next next batch of lights. A lot of people will try to prevent you from merging, but someone will always eventually let you through. It's called exit lane jumping. It's illegal but I highly doubt anyone gets pulled over for it.
Interesting! I've seen this happen a few times, though I've never witnessed something as extreme as a 10+ minute wait just for the off-ramp. I still maintain that it seems dangerous regardless of the situation, because while someone stops and tries to cut in line, the non-off-ramp lane they're stopped on can still keep moving, creating opportunities for collisions.
It's more common in larger metro areas - NYC, LA and Atlanta are infamous for it - but can happen anywhere depending on what is going on further down the exit lane (an emergency vehicle, car breakdown, etc).
Chicago's Lake Shore Drive/Belmont intersection has 3 stoplights you have to pass to get out. During rush hour it sometimes gets so backed up my bus has gotten stuck in it for a half hour. The first light is for the northbound on/off and a small side street that goes to some tennis/etc courts, the second for southbound on/off, and the side street adjacent to LSD splits in two so the third light is for the same street as the second light and is where you can finally enter it.
I think the issue is that the long-term decline in social trust—and the accompanying rise in surveillance, authoritarian enforcement, and costs/prices—happens too slowly for people to notice and associate with their own actions. If every time they left a cart out there was a new camera or scowling security officer on their next visit, they might notice and change their behavior. But as it is, they don’t notice their own contribution to the consequences that they so often complain about.
(Just so nobody misunderstands me, this is not to say that I want more cameras and security officers. Quite the opposite, which is why I don’t like casual antisocial behavior and petty crime.)
>But WHY do I do it? Why do I feel like I HAVE to do it? Am I actually experiencing any benefit in life over those who don't?
You start to skip the little things in life and it creeps up into the big things. "Do I have to return the shopping cart?" "Do I have to cook tonight?" "Do I have to shower today?" "Do I have to acknowledge that chatty neighbor and instead just walk past him?".
When your care starts to slip about participating in society, you start to disassociate with society. And I think times like these are where we need to care more than ever about participating.
A zipper merge is only applicable to a lane that is ending or has an obligation to merge. In exit lane jumping, the car is coming from a lane that does not end and has no obligation to merge. In fact, at their point of entry, you will notice a solid white separating the two lanes, indicating you cannot even legally merge.
There's a difference between zipper merging on a lane closure (which is what the article described) and what the person you are responding to described.
You are not supposed to block your lane of traffic because you didn't want to wait like everyone else.
No moral hazard, but there is a hazard to not brushing your teeth: cavities, and more explicitly, the financial cost of having cavities repaired. Even the best dental insurance plans in America are capped at $1500 - $3000 of benefits a year, usually less than a single root canal or crown.
But that does open the door to a very interesting question (far outside of the scope of this discussion): would people change their habits of brushing their teeth if that hazard didn't exist, e.g. your dental repairs were free? D:
One explanation I've heard that resonates with me is that we subconsciously feel as if we're playing a more complex and less obvious version of the prisoner's dilemma.
We intuitively understand that society experiences the greatest collective benefit when people generally cooperate. We also understand that while defecting (i.e. behaving in a selfish and anti-social way) might benefit us more as individuals, that's only true so long as others aren't also defecting. If they do, not only is society worse off but you personally are worse off as well than if everybody cooperated. And we understand that personally defecting leads to others doing the same.
Leaving your cart randomly in the parking lot, holding the train door open, or cutting across traffic may optimize your personal outcome, but the more people who behave like you the worse your grocery store parking lot experience gets, the more delayed your train is, and the longer you're stuck in traffic.
The nuance here is that modern societies are large enough that you can buy into the idea that your personal behavior does not influence the behavior of others in a way that will come back around to bite you. In a large metro area, what is the probability that the driver you cut off will be in a position to cut you off tomorrow? Ignoring the fact that society is smaller than you think when you look at sub groups like people who regularly drive on a certain road at a certain time, you have to consider second and third order effects. If cutting people off in traffic leads to more people cutting each other off in traffic, the impact spreads until it could easily come back around to your personal traffic experience with a few degrees of separation.
Fundamentally I think rational self-optimizing people realize that shitty personal behavior leads if only in a small way to the overall enshitification of society and that sooner or later this will come back around to negatively impact them personally. The people who engage in such behavior anyways aren't more rationally self-optimizing, they're either too stupid to see the connection or nihilistic enough to not care.
I don't understand this (and I didn't understand the point in the post).
When we discuss someone's net worth, we are specifically discussing their assets less their liabilities. We use it primarily to distinguish their purchasing power and credit-worthiness.
It is not a metric that is attempting to define their worth as a person. What standardized metrics could you possible use to measure that, and for what purpose would you use that metric?
If you're filling out a mortgage application in a Nordic country, are these hypothetical underpaid women and minorities considered more credit worthy regardless of their net worth and income?
> It is not a metric that is attempting to define their worth as a person.
You may not read it that way, but when you've never encountered the question before, the first time you see it being asked in the first place, it's comes across, not as an innocent question on a form that's just a reasonable part of a big process, but as a confrontation of a foreign culture that you've read and heard a lot about your whole life, only to be confronted by in that moment: What are you worth as a person?
That's not a common question to get asked. Okay, fine, the questionnaire is only asking as a business process thing, but the estimate is at about $10 million when broken down for parts, but at the point where someone's asking that question in the first place, you have to ask why are they asking?
Which you also point out,
> for what purpose would you use that metric?
The difference between worth and net worth is only one word, but like "guys" and "you guys", that one word makes a world of difference.
How would you define someone's worth as a person? It's because we don't talk about that at all, that even the question of net worth in the first place comes across as having a slight whiff of eugenics, because we have no other standardized measurements. Net worth is the only evaluation of how much any given individual a human is worth that has a magazine for it and list of all the high scores.
This doesn't seem to reflect the whole story or the gparent post about the word choice of "worth" instead of something closer to what you're describing. Trying to twist the point into credit risk also doesn't fit here.
To paraphrase gp, they found it shocking to have the word for a persons value to be the word used when describing how much money they have access to.
I've personally heard many people many times describe money and income as a way to measure either someone's value to society or how much society values them. This is very much in line with the gp - why would wealth have anything to do with your value as a person.
> “It is not a metric that is attempting to define their worth as a person.”
Yet that’s literally the word being used.
Imagine if a language called men “the better sex.” One could argue that it’s just a word and people don’t take it for its literal meaning. But you’d wonder why people go along with that. Don’t they notice what they’re saying? That’s the feeling I got from “person X is worth $Y” back when I first heard it.
Net worth is purely about assets minus liability. “How many dollars are attached to your tax identity and how many dollars of stuff can be taxed”
It has zero to do with the value of the the life of a person. You can conflate the two if you’d like, you’re picking on shortcut verbiage so we don’t say a paragraph of disclaimer text before talking about net worth.
I’m not conflating the two. I’m describing my experience encountering a culture that uses “worth” to mean the sum of a person’s material possessions. My own cultural background had primed me to think of these as entirely unrelated concepts.
You can argue it’s just a word, and that’s fine. There’s a whole another philosophical argument about if / how much words affect beliefs and actions.
One can be very wealthy and have zero net worth. Wealthy in family, security, etc. The net wealth of a toddler in perfect living conditions is almost infinite. Their net worth however, is 0.
They will wait and see if there's any deal heat.
Are you talking to other funds? You need to talk to as many funds as possible in a 2-3 week period to create leverage. Do not talk to a single or small number of funds in a process[0]. Best case they will snake the round at a discount, worst case you'll give up a ton of leverage and kill the process.
[0] unless you have close relationships with a stable of funds. you do not.
reply