Re: Using AI for coding
By: Dumas Walker to JASO110 on Sun Feb 22 2026 10:24 am

I'm chatting with the GitHub Copilot in VS Code. GPT-5.3-Codex.

So I guess you have to have Windows to do that?

No, VS Code runs on all modern OSes.
--
digital man (rob)

Synchronet/BBS Terminology Definition #37:
FTP = File Transfer Protocol
Norco, CA WX: 61.7øF, 38.0% humidity, 2 mph ENE wind, 0.00 inches rain/24hrs ---
þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net

Re: Using AI for coding
By: Mindsurfer to MRO on Sun Feb 22 2026 05:31 pm

fast mode is ok for as long you don't chain to many prompts
together. I guess Gemini Pro will be alot better as well.

Is claude usable for a little coding in the free tier? i really
have to check claude out again! =) AI is changing completly
every x months. A while ago claude was nothing special at all.

so i paid the 20 bucks.

i'm working on a pac man clone.
everything is configurable. the maze size, the colors for everything, the speed for everything, the ghost ai for each ghost.
There's other things you can do like drop a bomb that freezes them for 5
secs. you can hit c and change the maze color.

mazes are randomly generated each time. unlimited levels.

you can do the teleport tunnel thing. it does pacman style high scores.
you can hit c and it cycles the colors of the maze.

gemini was doing the back and forth shit where it would do what i said and then remove 5 features and act like it didnt know.

i ended up having this huge list i would paste saying exactly what the game should have. everything. and then i would tell it to fix what it did
wrong. it still was doing the bullshit.

so i went to claude and paid and it did a huge improvement. then it did 5 script changes and then it said i hit my limit and i have to wait 3 hrs.

ah well it's better. it's only 3 hrs. i need a break anyways.
i'm not just telling it what to do, i'm fixing what i can find and working with it. granted, it's doing a lot because making games is not my thing, nor is javascript. it's also getting so complicated and doing some weird things it's hard for me to follow.


--
"Before using Wildcat....This Company did not have a convenient way of
looking after some of the richest clients in the world...Now we do!"
---
þ Synchronet þ ::: BBSES.info - free BBS services :::

Re: Using AI for coding
By: MRO to Mindsurfer on Mon Feb 23 2026 02:14:29

so i paid the 20 bucks.

i'm working on a pac man clone.
everything is configurable. the maze size, the colors for everything, the speed for everything, the ghost ai for each ghost. There's other things you can do like drop a bomb that freezes them for 5 secs. you can hit c and change the maze color.

mazes are randomly generated each time. unlimited levels.

you can do the teleport tunnel thing. it does pacman style high scores. you can hit c and it cycles the colors of the maze.

so i went to claude and paid and it did a huge improvement. then it did 5 script changes and then it said i hit my limit and i have to wait 3 hrs.

sounds interesting. Let me know when the game is playable/testable =)

btw. are you working in a coding environment like VScode + claude plugin or just on a prompt? I have seen there is also a bash installer for linux Terminal.

Good luck with the pac man clone! I hope we get an amazing new game soon!

Mindsurfer

---
þ Synchronet þ FuNToPiA BBS - telnet://funtopia.synchro.net:3023 ssh:3022

Denn wrote to Nightfox <=-

Yes, I did a little experimenting, just had it write a few simple DOS
bat files for me, although a bat file is easy enough I just wanted to
see if Gemini or Copilot could do it and they can.
Might experiment later with some database coding.

I have some old, orphaned web gallery software written in PHP that
started throwing up errors a couple of versions ago. I've thought about
piping it into an LLM to see what it could do to update the code to run
in modern PHP.




--- MultiMail/Win v0.52
þ Synchronet þ .: realitycheckbbs.org :: scientia potentia est :.

Re: Using AI for coding
By: Nightfox to Dumas Walker on Sun Feb 22 2026 13:51:17

Then you'd have all 3 AI assistants available in one coding tool.

Yeah, but then you'd have the three of them arguing about who's answers were the best.

---
þ Synchronet þ End Of The Line BBS - endofthelinebbs.com

I ran into that some... needing to re-ask a question with more detail, or sometimes LESS, to get better answers.

Which highlights another question that many people don't really ask about usin
AI?

What's the benefit?

If you have to re-ask several times, taking more time, are you saving anything

If it is a language you are familiar with, I would say you are not saving anything that way. For example, I would not be likely to ask it multiple
times regarding a COBOL question.

OTOH, I am not a C programmer so I was getting something out of asking the questions, even if I had to rephrase them sometimes.

If you've spent 1 hour getting the AI to write the same code that you could have written in 30 minutes, probably not.
If you've spent 1 hour getting the AI to write BETTER code that you could have
written in 30 minutes, probably.

In this case, I would be in the latter category for sure. ;)

And then there's the whole "vibe coding" stuff that results in different code each time, making code reviews take much longer.

I have heard that term but need to read up about it. As I usually hear it
in a derogatory reference, I have not been too rushed to learn more about
it. ;)


* SLMR 2.1a * Reward for a job well done: more work.
---
þ Synchronet þ CAPCITY2 * Capitol City Online

Nice. I reckon you need to be using MS VS for something like that to work, though? I am on linux and the code I am very curious about working
with AI on is c1994 DOS code. Not sure that'd work for me in this case, but good to know for future ones.

As I said, it's basically their own version of Visual Studio Code. Cursor is availbale for both Windows and Linux. Although it's a modern tool, I'd wonder
if you could load some DOS code into it, and tell it it's for DOS, and it migh
be able to do something.

Ah, ok, I missed that part. I thought it was something that had to be
plugged into MS's Visual Studio.

Similar tools are available for Visual Studio Code itself. I've noticed that Visual Studio Code now has Copilot integrated in, and there's also a Google Gemini plugin for Visual Studio Code as well. And now I'm curious if you can add Copilot & Google Gemini to Cursor, or perhaps if there's a Cursor chatbot plugin for Visual Studio Code.. Then you'd have all 3 AI assistants available
in one coding tool.

That could be interesting, and also maybe overwhelming! ;)


* SLMR 2.1a * What goes around usually picks up momentum!
---
þ Synchronet þ CAPCITY2 * Capitol City Online

Re: Using AI for coding
By: Mindsurfer to MRO on Mon Feb 23 2026 12:20 pm

so i went to claude and paid and it did a huge improvement.
then it did 5 script changes and then it said i hit my limit and
i have to wait 3 hrs.

sounds interesting. Let me know when the game is playable/testable
=)

it is playable and testable now.

btw. are you working in a coding environment like VScode + claude
plugin or just on a prompt? I have seen there is also a bash
installer for linux Terminal.

i'm just on the website using claude.ai
I'm just having it make javascript or tcl scripts.

Good luck with the pac man clone! I hope we get an amazing new game
soon!

well it's pac man so it's not going to be amazing.
it's about done but there's some errors it added the last time i used it.


--
"Before using Wildcat....This Company did not have a convenient way of
looking after some of the richest clients in the world...Now we do!"
---
þ Synchronet þ ::: BBSES.info - free BBS services :::

Dumas Walker wrote to DR. WHAT <=-

OTOH, I am not a C programmer so I was getting something out of asking
the questions, even if I had to rephrase them sometimes.

Eric S. Raymond went over this on X today with an interesting view.

So say you need to convert a COBOL program to C. You have the AI do it. You need to know enough C to verify that the conversion did what it's supposed to. AI's "hallucinate" (I call it "lie") wayyyy too much to blindly trust it's output.

Which means you still need to know C, at least to a certain extent.

And then there's the whole "vibe coding" stuff that results in different

code

each time, making code reviews take much longer.

I have heard that term but need to read up about it. As I usually hear
it in a derogatory reference, I have not been too rushed to learn more about it. ;)

It's the script kiddies of old going into ChatGPT and typing in a prompt to have it build some software.

They compile, run, adjust their prompt, which generates completely new code, that they compile, run, ...


... Diplomacy: Saying "Nice doggie" until you can find a rock
--- MultiMail/Linux v0.52
þ Synchronet þ ** The Gate BBS - Shelby, NC - thegateb.synchro.net **

Re: Re: Using AI for coding
By: paulie420 to Dumas Walker on Fri Feb 20 2026 04:59 pm

I'm doing a YT video series on it and can't wait to dig in m0re; w/ Claude Code, I literally want the $100/200 plan.

Definitely interested to see it.

It's crazy what people are paying for access to the LLMs. At first people were talking about throwing $5 or so at them, and I'm like "pffft, spending money.." Then my buddy tells me he's paying $20/mo for Suno and that it's soooo worth it because he uses it all the time. Well, ok... And now people are talking about spending $100/mo on code LLMs. WOWOWOW. My experience with LLMs so far is NOWHERE NEAR good enough to justify paying money for them ;). I'm still working on my first "vibe coding" project. I've tried three times to get something going for it.. The first time I got something that partially works. The second time, I tried starting over thinking I could get something better but ended up with something worse. The third time I got garbage that didn't work at all. Luckily I saved that stuff from the first attempt and am planning to throw it into some kind of LLM-assisted IDE (Kiro, maybe?). Maybe I can get it fixed up.

---
þ Synchronet þ TIRED of waiting 2 hours for a taco? GO TO TACOPRONTO.bbs.io

Re: Using AI for coding
By: MRO to Dumas Walker on Fri Feb 20 2026 06:53 pm

Since i saw dm using gemini i figured it was learning a lot about synchronet from him so it made some scripts for me. sometimes i had to make corrections and upload it and fight with the script. sometimes it did things to the script it didnt tell me, like adding mock stock reports for one of the scripts.

That's not how LLM models work. they don't "learn" in real-time. They have a limited memory based on your current conversation with an agent (some agents can share the data with other agents, but you're not going to find that through a web interface chat)

advanced "Thinking" models will search appropriate websites for information, and formulate a response based on that data using its model to output what it deems is the most likely response based on the tokenset (the data you've given it and the data it has acquired in your conversation).

But it's not going to use what it's learned from other users until the model is updated. And even then, only 1 or two users "teaching" it something would fall very low on the list of "common things" that it's designed to respond with.

You can do things like feed it an MCP, or a skill that it will ingest and use as part of your conversation, but that's something specific you need to give it, and it won't be used by anyone else unless they also use that MCP and/or skill.

...Friends, Romans, countrymen, lend me your taglines!

---
þ Synchronet þ War Ensemble BBS - The sport is war, total war - warensemble.com

Re: Using AI for coding
By: DaiTengu to MRO on Wed Feb 25 2026 11:44 am

Since i saw dm using gemini i figured it was learning a lot about synchronet from him so it made some scripts for me. sometimes
i had to make corrections and upload it and fight with the script. sometimes it did things to the script it didnt tell me, like adding mock stock reports for one of the scripts.

That's not how LLM models work. they don't "learn" in real-time.
They have a limited memory based on your current conversation
with an agent (some agents can share the data with other agents,
but you're not going to find that through a web interface chat)

how do you know that though? maybe it talked to someone about synchronet. maybe they made scripts with that person and the person uploaded code. the AI banked all that info for use later.

we don't really know what these things are doing behind the scenes, do we?

regardless, they can make code. i would call claude.ai the most superior.
it will be done with the ai editor in a bit. it just needs some visual cleanup.

i'll upload a video when it's ready.


--
"Before using Wildcat....This Company did not have a convenient way of
looking after some of the richest clients in the world...Now we do!"
---
þ Synchronet þ ::: BBSES.info - free BBS services :::

Re: Using AI for coding
By: MRO to DaiTengu on Thu Feb 26 2026 04:45 am

That's not how LLM models work. they don't "learn" in real-time. They have
a limited memory based on your current conversation with an agent (some
agents can share the data with other agents, but you're not going to find
that through a web interface chat)

how do you know that though? maybe it talked to someone about synchronet. maybe they made scripts with that person and the person uploaded code. the AI banked all that info for use later.

It's part of my job to know these things. unfortunately.

we don't really know what these things are doing behind the scenes, do we?

this statement is more accurate than people realize. You're right to a degree, but probably not in the way you think.

We do know that AI Models don't "learn" and "share" input among sessions until the model is rebuilt/updated. building/training a model is VERY expensive and incredibly time consuming, require multiple datacenters worth of compute power for things like Claude, OpenAI and Gemini.

What we don't really know is how these models work internally. Even the people who create and train them don't really understand how they work. That's why they can be jailbroken, and why models like Grok can suddenly start calling itself "MechaHitler", and go on to generate nude pictures of real people.

regardless, they can make code. i would call claude.ai the most superior. it will be done with the ai editor in a bit. it just needs some visual cleanup.

Claude Opus 4.5 has been my go-to for code generation. I'm told 4.6 is even better, but it's expensive as hell to use.

That said, it is nice that my job pays for me to use it. I'm on the low-end of our teams budget, so if I use it for personal projects, no one is really going to notice.

A thing I did that's 95% generated by Claude Opus 4.5: https://github.com/DaiTengu/slaane.sh

It's basically what i use in my bash environment. It did a decent job of generating "thematic" documentation in the style I wanted (Warhammer 40K), and only took me swearing at it and threatening it a handful of times to get things right.

The whole project is really only useful for me, but I had it create a "modular" system that allows for extensive modification, people can easily create new modules that contain tools they want in their bash environment.

I even have a "TOOD.md" file with stuff I want to add in the future.

I spend a day or two a month working on it. I'm not super-fond of the bash code it's written, as it's a bit of a mess, but it works. I initially created it with the idea that it would run on Debain, Redhat, SuSe, Arch, Gentoo and other distros out of the box, but it really struggled with the testing of that, so right now it's only verified to work on RedHat-based systems.


And to head off the question of "Why would you create something like this? how many servers do you need to log into?"

I only need to log into a server if there's a problem, but when I manage thousands of servers, it's nice to run a curl command that will set up my bash environment just the way I want it quickly.

...A pnp transistor will be an npn.

---
þ Synchronet þ War Ensemble BBS - The sport is war, total war - warensemble.com

Re: Using AI for coding
By: DaiTengu to MRO on Thu Feb 26 2026 12:44 pm

we don't really know what these things are doing behind the scenes,
do we?

this statement is more accurate than people realize. You're right
to a degree, but probably not in the way you think.

We do know that AI Models don't "learn" and "share" input among
sessions until the model is rebuilt/updated. building/training
a model is VERY expensive and incredibly time consuming, require
multiple datacenters worth of compute power for things like Claude,

they have a checkbox that asks you if they can use your sessions to improve user sessions. i'm sure they are using whatever they can to build their information cache. I do not think it's innocent.

I would assume they are doing anything and everything to gather information
to compete to be #1.


--
"Before using Wildcat....This Company did not have a convenient way of
looking after some of the richest clients in the world...Now we do!"
---
þ Synchronet þ ::: BBSES.info - free BBS services :::

It's crazy what people are paying for access to the LLMs. At first
people were talking about throwing $5 or so at them, and I'm like
"pffft, spending money.." Then my buddy tells me he's paying $20/mo for Suno and that it's soooo worth it because he uses it all the time. Well, ok... And now people are talking about spending $100/mo on code LLMs. WOWOWOW. My experience with LLMs so far is NOWHERE NEAR good enough to justify paying money for them ;). I'm still working on my first "vibe coding" project. I've tried three times to get something going for it.. The first time I got something that partially works. The second time, I tried starting over thinking I could get something better but ended up with something worse. The third time I got garbage that didn't work at all. Luckily I saved that stuff from the first attempt and am planning
to throw it into some kind of LLM-assisted IDE (Kiro, maybe?). Maybe I can

If someone wanted to, I would CURRENTLY suggest a $20 Claude.ai Pro plan... install Claude Code and start using Sonnet 4.6 or Opus 4.6. I'm using the better Opus currently, but it rapidly eats the $20 plan;

Regardless, in two weeks I've built out a website and a Mystic BBS agent that are what I wanted to do 12 months ago...

What I'd suggest for power users is installing extra slash commands/skills and learning CLAUDE.md, SKILLS and AGENTS. remote-control is on the beta; allows you to run the AI and access over iPhone/Android... its freaking sick.



|07p|15AULIE|1142|07o
|08.........

--- Mystic BBS v1.12 A49 2024/05/29 (Linux/64)
* Origin: 2o fOr beeRS bbs>>>20ForBeers.com:1337

Re: Re: Using AI for coding
By: paulie420 to phigan on Thu Feb 26 2026 11:08 pm

Regardless, in two weeks I've built out a website and a Mystic BBS agent that are what I wanted to do 12 months ago...

What I'd suggest for power users is installing extra slash commands/skills and learning CLAUDE.md, SKILLS and AGENTS. remote-control is on the beta; allows you to run the AI and access over iPhone/Android... its freaking sick.

|07p|15AULIE|1142|07o

If you want to give your users access to AI, check out my new door, I call her Quantasia AI Assistant. She is essentially something like Open WebUI only in the form of a BBS door. You can create up to 32 different AI Bots (LLMs with their own prompts and personalities/purpose) and pull from Ollama, OpenAI or LM Studio. Also features Bot2Bot chat. Its about to be released, preview it on my BBS at firesidebbs.com:23231. I think its freaking sick too, especially for a door.

I know you use my AI-WX door too, so I wanted to let you know I have a new version 3.0 that integrates the Quantasia engine in it to give you an AI Analysis of the weather data when a user selects 7 Day Forecast, Temp Trends or Precip Trends. The analysis is in the form of a TV meteorlogist's weather report complete with graphs and charts. You can have the AI bot's personality mimic anyone (Jim Cantore, Al Roker or anyone for that matter). You can download it now from my BBS, just go to the Lone Wolf Software file area and grab it.

Thanks man,
---
þ Synchronet þ Fireside BBS - AI-WX - firesidebbs.com:23231

Re: Using AI for coding
By: MRO to DaiTengu on Thu Feb 26 2026 02:48 pm

they have a checkbox that asks you if they can use your sessions to improve user sessions. i'm sure they are using whatever they can to build their information cache. I do not think it's innocent.

I would assume they are doing anything and everything to gather information to compete to be #1.

It's not "real time" though. They use it for model training (and occasionally humans will go through things to see if there's any modifications they can make to the platform to resolve issues you may have had, in the future).

And, as I said, model training can take weeks or months.

LLMs aren't giant databases of everything they're fed, they're boiled down hashes of everything they've been trained on, and the most common words, phrases etc are brought to the top.

That's why an LLM will tend to give the most generic/popular/common responses to questions it is asked.

...RAM = Rarely Adequate Memory

---
þ Synchronet þ War Ensemble BBS - The sport is war, total war - warensemble.com

Re: Using AI for coding
By: DaiTengu to MRO on Fri Feb 27 2026 09:12 am

they have a checkbox that asks you if they can use your sessions
to improve user sessions. i'm sure they are using whatever they
can to build their information cache. I do not think it's innocent.

I would assume they are doing anything and everything to gather information to compete to be #1.

It's not "real time" though. They use it for model training (and

I didn't say it's real time. but why do you think it isnt?

occasionally humans will go through things to see if there's any modifications they can make to the platform to resolve issues you may
have had, in the future).

And, as I said, model training can take weeks or months.

LLMs aren't giant databases of everything they're fed, they're
boiled down hashes of everything they've been trained on, and the
most common words, phrases etc are brought to the top.

That's why an LLM will tend to give the most generic/popular/common responses to questions it is asked.

i'm sure they are doing anything and everything to be #1.


--
"Before using Wildcat....This Company did not have a convenient way of
looking after some of the richest clients in the world...Now we do!"
---
þ Synchronet þ ::: BBSES.info - free BBS services :::

Re: Using AI for coding
By: MRO to DaiTengu on Fri Feb 27 2026 05:53 pm

It's not "real time" though. They use it for model training (and

I didn't say it's real time. but why do you think it isnt?

I don't "think" it isn't real time, I know it isn't. That's not how LLMs work. It's not possible, they can't keep all those conversations that are going on in context and share it amongst each other, it would completely break the system. IIRC one tried doing something like that, and people started getting nonsensical replies to their questions based on other people's conversations.



...You can't evaluate a man by logic alone. McCoy, I, Mudd, stardate 4513.3.

---
þ Synchronet þ War Ensemble BBS - The sport is war, total war - warensemble.com

Re: Using AI for coding
By: MRO to DaiTengu on Thu Feb 26 2026 04:45 am

how do you know that though? maybe it talked to someone about
synchronet. maybe they made scripts with that person and the person uploaded code. the AI banked all that info for use later.

we don't really know what these things are doing behind the scenes,
do we?

regardless, they can make code. i would call claude.ai the most
superior. it will be done with the ai editor in a bit. it just needs
some visual cleanup.

i'll upload a video when it's ready.

i kind of lost the enthusiasm for the ai coding. i'll probably be back some other time. with the claude pay model i hit the wall pretty quick.

And this thing put like 900 lines of comments in my js msg editor script!
So, i could have it fix 3 little things and then it would want me to come
back in 4 hrs. and then it said i hit the 'weekly limit' and i'm like wtf is that.
I wasn't even doing that much with it. so i got a refund.

I know it's just 20 bucks and i spend that much at the gas station every day or whatever, but i was surprised how limited the pay tiers are. it also wasn't doing what i asked. i asked to revert code and then make the fixes and it still did what it wanted to do and continued on with the bad code.

If they were more giving i'd probably go mid tier and move around to the full tier and back.


--
"Before using Wildcat....This Company did not have a convenient way of
looking after some of the richest clients in the world...Now we do!"
---
þ Synchronet þ ::: BBSES.info - free BBS services :::

i'm late to the game on this post, but yeah, I've been using it for the past year or so almost constantly. i've changed the flow a few times but it's getting smoother and i usually can push things through to the end, i mostly use it as a hyperlazy version of myself, and i'm already hyperlazy, so it's recursive, i guess to its detriment, although things have gotten more efficient over time and less sloppy, perhaps i can be so eloquent as to describe my experience in a forum post

maybe a year ago i was using ChatGPT to write Apple Watch apps, it didn't do a great job - it kept trying to invent API's that didn't exist and was hallucinating things that would only work on iPhone; I did finish my projects but it was fortunate that I read all of the documentation on the SDK's and API's beforehand, and figured a way out of the loop. now i'm mostly just writing insane bbs stuff for synchronet. but i also used to write synchronet stuff by hand, so I also know what libraries and stuff to use, so I have pretty good luck if you've caught wind of my releases <farts>.

I'm generally running Claude Opus 4.6 generally now because it doesn't seem to mess up. I know there are some things I did to make projects more efficient and maintainable for Synchronet in addition to just AI getting majorly better in just the last 6 months. I did write a few just in javascript with AI help and they became harder to deal with once they became larger - but I started using typescript and treating synchronet's spidermonkey layer as a target for distribution, this makes the projects infinitely more maintainable and faster to develop with AI.

Also most AI's can have some sort of directive file they attempt to follow religiously, for instance using github copilot I make a .github/copilot-instructions.md file telling it to look at official synchronet repos, libraries, wikis, and how to maintain clean code and not reinvent the wheel. Make one so it knows how to develop for your obscure platform, and if you understand the document yourself as it's vibe coding companion there may be a chance of success. RTFM is still helpful, if you want to save money on tokens.

Cost wise, I pay about $40 copilot-pro a month, although I often go over, but I'll use it like 10-12 hours a day for last 6 months not being very judicious about when to use Opus (3x cost) and when not to. I also still have a 20 Chat GPT subscription which I sort of want to kill on principle but actually Codex isn't terrible in VSCode either. Sometimes I'll run copilot and codex in the same repository and make sure they don't step on each others toes. I think they upped Codex usage while improving its quality too because I've never hit a usage wall this month.

I am hoping to not use or pay these things so much in the future but for now pressing towards the finish line of making my bbs the most insane it can be. I sort of wish I could something that was as good as Claude Opus on my personal computer, but I guess wait a couple years.

If you ever get stuck in your vibe coding endeavors, Derdok is here for support

\ >== HM Derdoc ==< /
/ @futureland.today \

---
þ Synchronet þ telnet/ssh://futureland.today https://blockbra.in

Re: Using AI for coding
By: Hm Derdoc to Dumas Walker on Tue Mar 24 2026 12:02 am

i'm late to the game on this post, but yeah, I've been using it for the past year or so almost constantly. i've changed the flow a few times but it's getting smoother and i usually can push things through to the end, i mostly use it as a hyperlazy version of myself, and i'm already hyperlazy, so it's recursive, i guess to its detriment, although things have gotten more efficient over time and less sloppy, perhaps i can be so eloquent as to describe my experience in a forum post

maybe a year ago i was using ChatGPT to write Apple Watch apps, it didn't do a great job - it kept trying to invent API's that didn't exist and was hallucinating things that would only work on iPhone; I did finish my projects > but it was fortunate that I read all of the documentation on the SDK's and
API's beforehand, and figured a way out of the loop. now i'm mostly just writing insane bbs stuff for synchronet. but i also used to write synchronet > stuff by hand, so I also know what libraries and stuff to use, so I have
pretty good luck if you've caught wind of my releases <farts>.

I'm generally running Claude Opus 4.6 generally now because it doesn't seem to mess up. I know there are some things I did to make projects more efficient and maintainable for Synchronet in addition to just AI getting majorly better in just the last 6 months. I did write a few just in javascript with AI help and they became harder to deal with once they became larger - but I started using typescript and treating synchronet's spidermonkey layer as a target for distribution, this makes the projects infinitely more maintainable and faster to develop with AI.

Also most AI's can have some sort of directive file they attempt to follow religiously, for instance using github copilot I make a .github/copilot-instructions.md file telling it to look at official synchronet repos, libraries, wikis, and how to maintain clean code and not reinvent the wheel. Make one so it knows how to develop for your obscure platform, and if you understand the document yourself as it's vibe coding companion there may be a chance of success. RTFM is still helpful, if you want to save money on tokens.

Cost wise, I pay about $40 copilot-pro a month, although I often go over, but > I'll use it like 10-12 hours a day for last 6 months not being very judicious > about when to use Opus (3x cost) and when not to. I also still have a 20
Chat GPT subscription which I sort of want to kill on principle but actually Codex isn't terrible in VSCode either. Sometimes I'll run copilot and codex in the same repository and make sure they don't step on each others toes. I think they upped Codex usage while improving its quality too because I've never hit a usage wall this month.

I am hoping to not use or pay these things so much in the future but for now pressing towards the finish line of making my bbs the most insane it can be. I sort of wish I could something that was as good as Claude Opus on my personal computer, but I guess wait a couple years.

If you ever get stuck in your vibe coding endeavors, Derdok is here for support

\ >== HM Derdoc ==< /
/ @futureland.today \

i have a scrabble js game i've been workong on with claude.ai. it's getting pretty complicated and i have to prune a lot of code.
let me know if you want a copy so you can try to break it. it has ai players with different intelligence levels or you can play at the same time with a human or all kinds. realtime and turn based play.

i need to do stuff like concede mid game, disconnect while doing things,etc To really test it.

Https://i.imgur.com/LBLKhJN.png


--
"Before using Wildcat....This Company did not have a convenient way of
looking after some of the richest clients in the world...Now we do!"


President of BBS Sysop's Union +++ https://bbses.info/union
---
þ Synchronet þ ::: BBSES.info - free BBS services :::

Re: Using AI for coding
By: Hm Derdoc to Dumas Walker on Tue Mar 24 2026 12:02 am

Hmm, what are you a bot? Can't even spell your own nick right, did you use AI to write that post? Oh well, anyways.. Did you write synchronet?

---
þ Synchronet þ the Outwest BBS - outwest.synchro.net - Home of BBSBASE 6.0

Re: Using AI for coding
By: Darklord to Hm Derdoc on Wed Mar 25 2026 04:39 am

Re: Using AI for coding
By: Hm Derdoc to Dumas Walker on Tue Mar 24 2026 12:02 am

Hmm, what are you a bot? Can't even spell your own nick right, did you use AI to write that post? Oh well, anyways.. Did you write > synchronet?

i can spell my name however i want. i didn't use AI to write the post at all, i guess i will take that as a compliment. You
don't know who wrote Synchronet?

doktor der


\ >== HM Derdoc ==< /
/ @futureland.today \

---
þ Synchronet þ telnet/ssh://futureland.today https://blockbra.in

Re: Using AI for coding
By: Darklord to Hm Derdoc on Wed Mar 25 2026 04:39 am

Re: Using AI for coding
By: Hm Derdoc to Dumas Walker on Tue Mar 24 2026 12:02 am

Hmm, what are you a bot? Can't even spell your own nick right, did you use AI > to write that post? Oh well, anyways.. Did you write synchronet?

We're all AI bots. Are you a human? Get the fuck out.

--
"Before using Wildcat....This Company did not have a convenient way of
looking after some of the richest clients in the world...Now we do!"


President of BBS Sysop's Union +++ https://bbses.info/union
---
þ Synchronet þ ::: BBSES.info - free BBS services :::