Your way of thinking of constants and naming seems a bit "worst case" thinking?
And if it were to grow I would probably look into making a FlatFileParser which then encapsulated the constants, if you need to access the mostly random indexes outside of parsing I question why?
An enum could be a partial answer inside of this parser if it then keeps getting more and more correlated constants
Well then you will hate to know you define AI wrong. It's just artificial intelligence, it does not mean that the intelligence is particularly smart. AI really just means a computer imitating/performing something that we generally consider a challenge of the mind. A calculator, maze solver, sudoku solver, chess ai, video game NPC, and LLM all count under this. Just as bacteria counts as biological intelligence
For europæisk kan du evt prøve Mistral, ellers ville kinesisk deepseek v4 nok være en god billig en for bare at prøve det. Har selv codex/openai privat og GitHub copilot på arbejdet
Lokale modeller er faktisk blevet gode, men vi snakker stadig at man skal have dedikeret highend gear for rigtig at bruge det (noget der let kan koste 100k kroner).
Hvis du har en enkel god gpu (4090, 5090 eller Mac med meget integrated ram) kan du rimlig hurtigt prøve en af de mindre open source modeller som qwen 3.6 (27b eller 35b 3b expert) og se om den er god nok for dig, ollama (for at køre modellen) + pi.dev (som din "claude code") er nok hurtigst i opsætning
One must imagine the elevator happy
Also known as an American™️
Gives you (up to) 52 shots at any plans, perfect
Don't forget insert disk 2, flip to backside of disk 1, now disk 3
In the needy spectrum?
It took the llm 11 days to generate the code
It doesn't seem to say anywhere anyone reviewed it in any time
tbf high prevalence is a double edged sword, it also means it's trained on a lot of poorly written code which it will slide into
100 people? 3 packs
Your way of thinking of constants and naming seems a bit "worst case" thinking?
And if it were to grow I would probably look into making a FlatFileParser which then encapsulated the constants, if you need to access the mostly random indexes outside of parsing I question why?
An enum could be a partial answer inside of this parser if it then keeps getting more and more correlated constants
Well then you will hate to know you define AI wrong. It's just artificial intelligence, it does not mean that the intelligence is particularly smart. AI really just means a computer imitating/performing something that we generally consider a challenge of the mind. A calculator, maze solver, sudoku solver, chess ai, video game NPC, and LLM all count under this. Just as bacteria counts as biological intelligence
VERY NICE
This is just having constants like
final int FLATFILE_DATE_INDEX = 2with more steps and boilerplate?
That's just letting it run any command it wants with more steps
For europæisk kan du evt prøve Mistral, ellers ville kinesisk deepseek v4 nok være en god billig en for bare at prøve det. Har selv codex/openai privat og GitHub copilot på arbejdet
Lokale modeller er faktisk blevet gode, men vi snakker stadig at man skal have dedikeret highend gear for rigtig at bruge det (noget der let kan koste 100k kroner). Hvis du har en enkel god gpu (4090, 5090 eller Mac med meget integrated ram) kan du rimlig hurtigt prøve en af de mindre open source modeller som qwen 3.6 (27b eller 35b 3b expert) og se om den er god nok for dig, ollama (for at køre modellen) + pi.dev (som din "claude code") er nok hurtigst i opsætning
Ved mindre man timer det rigtigt og sidder med 1% lige før det hele stiger
Og her var jeg bange for at toge ville blive nedprioriteret...
It just seem like a weird place to go on fire
Gotta love Pascal's mugger