I worked years as backend and desktop software programmer, then in gamedev and now back to SaaS development mostly backend. I didnt have much success with agentic coding with "agents", but had a great success with LLM code generation while keeping all the code in context with Google Gemini.
For gamedev you can really build quite complex 2D game prototype in Pygame or Unity rapidly since 20-50KLOC is enough for a lot of indie games. And it allow you to iterate and try different ideas much faster.
Most of features are either one-shots doing all changes across codebase in one prompt or require few fixing prompts only.
It really helps to isolate simulation from all else with mandatory CQRS for gamestate.
It also helps to generate markdown readmes along the way for all major systems and keep feature checklists ih header of each file. This way LLM dont lose context ot what is being generated.
Basically I generated in 2-3 weeks projects that would take 2-3 months to implement in a team simply because there is much less delay between idea of feature and testing it in some form.
Yes - ocassiinally you will fail to write proper spec or LLM fail to generate working code, but then usually it means you revert everything and rewrite the specification and try again.
So LLMs of today are certainly suitable when "good enough" is sufficient. So they are good for prototyping. Then if you want better architecture you just guide LLM to refactor complete code.
LLMs also good for small self contained projects or microservices where all relevant information fits into context.
There is "international law", but no "international police". So the law only as good as long as everyone follow it - so not very good either.
Then you have major power like Russia that constantly abuse it while trying to cover its ass with "internatiomal law" once they themself get hit. This is whem you get system is broken.
I get it suck for people inside the US to see democracy dismantled, but honestly I dont mind Trump to deal with dictators and repressive regimes.
Putin, friends and alike put a lot of effort to prove that only power should matter so its a good irony to see Trump dealing with them their way.
In a war zone any large high power jammer will be like supernova in the darkness visible for detectors from tens of kilometers away. So its gonna be immediately destroyed.
Iran protesters cant find or destroy jammers though.
Does it really matter what official statements say if a country have stock exchange, group of billionares and market economy?
Iived there for quite some time to understand its likely less regulated and government controlled than many ex-USSR countries (im from russia).
Of course politically Vietnam is what it state it is compared to e.g Indonesia that I now explore. But economically it's the same capitalism as everywhere else.
Lots and lots of people work for much less or for free on whatever they like.
Problem is that doing "boring" parts of open source project maintenance is not very exciting for many top tier developers so it should pay at least competetively for experience or people will just burn out.
And while you can obviously fund a team of 20 on $1M/year outside of US whatever said team will manage to keep up to the level of quality is another question.
I do think about this in the context of other tech companies, the "bidirectionality of enforcement", or whatever you want to call it
Let's say you have Facebook, which is notorious for banning people yet never seems to ban the things people report that should be banned. That's a real life example, but take any hypothetical company
If someone posts x bad thing and doesn't get banned, do we immediately take our torches and storm the premeses to protest? Maybe, maybe not; "look, scale is hard" (and sometimes calls to remove things outright get politicized, as seen in the last few years, so sometimes it's a tricky line)
That would be... not fine, but more fine than it is now. The lack of fairness in the bidirectionality ensures that you, Joe Schmoe, get a month ban for calling someone a jerk while the most egregious hate or racism or... anything... gets a quick check followed by This Does Not Violate Our Community Guidelines
(And of course because these services are monopolies, well, too bad, you just have to suffer. Hope you don't need the information from that Facebook page, because Facebook will tend to make it borderline impossible to view something public without an account)
I think companies like Google dont even try like they are "Too Big to be Regulated".
Facebook is much worse because everyghing on there is user gemerated. Any small company would be just crushed by governments if they would have similar issues.
I think they are similar to FedEx. FedEx knows that millions of packages per day are transporting illegal goods, any bad enough accident shows it. However, FedEx would absolutely go bankrupt if they tried to open every package and make sure the contents were good. At the end of the day, that's the government's job.
If the DEA and ATF wants to staff every shipping hub with people checking every package, that's fine by them (though admittedly it would hurt revenues).
For Google and Facebook and all the other user-content sites, it's just impossible to actually, fully uphold the law themselves, so their best bet is just to try to make it a pleasant experience for the users and leave upholding the law to the upholders of the law.
To build things you also need a lot of people with education, know how and experience. You cant just bring low-wage workforce and expect to compete with China.
Let alone that to provide same quality of living to average chinese worker as they have in China their salary in US will have to grow 5-6 times.
> Let alone that to provide same quality of living to average chinese worker as they have in China their salary in US will have to grow 5-6 times.
That's Baumol's Cost Disease, a symptom of not allowing massive amounts of workers in. The solution is to let them in, something like how UAE or Hong Kong does it but even more.
Most people don't want to do that. That's a choice they made, but they must live with the consequences and stop complaining.
> To build things you also need a lot of people with education, know how and experience.
Let those people in too as guest workers. There's millions of people that know how to build housing, do surgery, and so on, but they're not allowed to do it.
For gamedev you can really build quite complex 2D game prototype in Pygame or Unity rapidly since 20-50KLOC is enough for a lot of indie games. And it allow you to iterate and try different ideas much faster.
Most of features are either one-shots doing all changes across codebase in one prompt or require few fixing prompts only.
It really helps to isolate simulation from all else with mandatory CQRS for gamestate.
It also helps to generate markdown readmes along the way for all major systems and keep feature checklists ih header of each file. This way LLM dont lose context ot what is being generated.
Basically I generated in 2-3 weeks projects that would take 2-3 months to implement in a team simply because there is much less delay between idea of feature and testing it in some form.
Yes - ocassiinally you will fail to write proper spec or LLM fail to generate working code, but then usually it means you revert everything and rewrite the specification and try again.
So LLMs of today are certainly suitable when "good enough" is sufficient. So they are good for prototyping. Then if you want better architecture you just guide LLM to refactor complete code.
LLMs also good for small self contained projects or microservices where all relevant information fits into context.