If you wouldn't mind answering a question for me, it's one of the main things that has made me not add claude in vscode.
I have a custom 'code style' system prompt that I want claude to use, and I have been able to add it when using claude in browser -
```
Beautiful is better than ugly. Explicit is better than implicit.
Simple is better than complex. Complex is better than complicated.
Readability counts.
Special cases aren't special enough to break the rules. Although practicality beats purity.
If the implementation is hard to explain, it's a bad idea. If the implementation is easy to explain, it may be a good idea.
Trust the context you're given. Don't defend against problems the human didn't ask you to solve.
```
How can I add it as a system prompt (or if its called something else) in vscode so LLMs adhere to it?
Same. AI is a good tool to use as a sounding board and conversation partner.
I only access claude and others using my browser - I give it a snippet of my code, tell it what exactly I want to do and what my general goal is, then ask it to give me approaches, and their pros and cons.
Even if someone wants to use AI to code for them, its still better to do the above as a first step imo. A sort of human in the loop system.
> It adds velocity but velocity early on always steals speed from the future. That's been the case for languages, for frameworks, for libraries, it's no different for AI.
Completely agree. I'm seeing this in my circle and workplace. My velocity might be a tad bit slower than the rest of my peers when you compare it per ticket. But my long tern output hasn't changed and interestingly, neither has anyone else's.
As an aside, I like your system of completely removing autocomplete unless you need it - may be something like that would finally get me to enable AI in my IDE.
A playground for the zon format is great, but it would be amazing to see a few examples where zon has already been integrated into the LLM and see its responses to user queries. It doesn't even need to be a playground (as that becomes costly quickly), just some examples for the user to see how the black box will work when zon is integrated.
I have attached the script and the logs comparing how ZON performs well everytime compared to TOON. and how ZON-FORMAT has a feature of Eval to get rid of LLM Hallucination.
I just created separate profiles for different stuff. Work is all on chrome anyways due to google integration, so all thats left in my random browsing.
Ordered in what I use the most -
Fanfics, novels - profile 1.
Netflix, others - profile 2.
General browsing - profile 3.
Somehow this became one of favorite posts ever shared on this website.
Its give a view into one of the most common thing used for entertainment in households worldwide now (specially if you consider phones and tablets a part of this). As a guy born in 1990s, I have never even made the connection - television itself is so ingrained into nearly all households.
I kind of like how good storytelling also relates to the article. A black person showing all the stereotyped black mannerisms in media is fine, but great storytelling builds a character whole. Adding a small thing like liking spicy indian food adds so much to the character.
Yeah. I'm all for sprints to start focusing on optimizations and bug smashing instead of just barreling down the road of profitability while the middle managers hold a gun to your head.
I would go against the grain and say that LLMs take power away from incredibly rich people to shape mass preferences and give to the masses.
Bot armies previously needed an army of humans to give responses on social media, which is incredibly tough to scale unless you have money and power. Now, that part is automated and scalable.
So instead of only billionaires, someone with a 100K dollars could launch a small scale "campaign".
"someone with 100k dollars" is not exactly "the masses". It is a larger set, but it's just more rich/powerful people. Which I would not describe as the "masses".
I know what you mean, but that descriptor seems off
I guess my non-management / non-business side is show here, but how can it be that much?? I still remember I designed a fairly simple cron job that took database backups when I was a junior developer.
It gets even easier now that you have cheap s3 - just upload the dump to s3 every day and set the s3 deletion policy to whatever is feasible for you.
I am not an expert here but I am currently researching for a planned project.
For backups, including Postgres, I was planning on paying Veeam ~$500 a year for a software license to backup the active node and Postgres database to s3/r2. Standby node would be getting streaming updates via logical replication.
There are free options as well but I didn’t want to cheap out on the backups.
It looks pretty turnkey. I am a software engineer not a sysadmin though. Still just theory as well as I haven’t built it out yet
Taking database backups is relatively simple. What differentiates a good solution is the ease of restoring from a backup. This includes the certainty that the restored state would be a correct point-in-time state from the past, not an amalgamation of several such states.
How much were you paid as a jr developer, and how long did it take you to set up? Then round up to mid-level developer, and add in hardware and software costs.
That's a deflection. The question isn't about a developer's salary; it's about the fundamental difference between a one-time investment and a permanent cost.
Either way: 1 day of a mid-level developer in the majority of the world (basically: anywhere except Zurich, NYC or SF) is between €208 and €291. (Yearly salary of €50-€70k)
A junior developer's time for setup and the cost of hardware is practically a one-off expense. It's a few days of work at most.
The alternative you're advocating for (a recurring SaaS fee) is a permanent rent trap. That money is gone forever, with no asset or investment to show for it. Over a few years, you'll have spent tens of thousands of dollars for nothing. The real cost is not what you pay a developer; it's what you lose by never owning your tools.
Time and physical labor are an amazing way to process grief.
There is a hindu ceremony in India where we build a funeral pyre and light it. Before that, there are prayer ceremonies that happen over a few days (I think). I've always felt that it was designed this way (way back when) to give time and space to the family to grieve. The lighting of the funeral pyre seems like letting go somewhat.
If you wouldn't mind answering a question for me, it's one of the main things that has made me not add claude in vscode.
I have a custom 'code style' system prompt that I want claude to use, and I have been able to add it when using claude in browser -
``` Beautiful is better than ugly. Explicit is better than implicit. Simple is better than complex. Complex is better than complicated. Readability counts. Special cases aren't special enough to break the rules. Although practicality beats purity. If the implementation is hard to explain, it's a bad idea. If the implementation is easy to explain, it may be a good idea.
Trust the context you're given. Don't defend against problems the human didn't ask you to solve. ```
How can I add it as a system prompt (or if its called something else) in vscode so LLMs adhere to it?
reply