/home/chiefgyk3d
Region: US
Friday 23 October 2026 21:56:02 GMT
12351
718
17
28
Music
Download
Comments
Dust In Consoles :
tutorial please
2026-10-23 22:39:29
8
Luke :
I don’t do agentic because I only pay for the $20 plan. One wrong opus plan and that’s gone
2026-10-24 04:23:22
5
Wreck It Mech :
OpenCode agents. I rarely hit my max on CoPilot Max, unless I'm actively trying to burn tokens. Personas and skills really matter!
2026-10-24 16:20:06
0
nmbr5 :
my issue is I'm running local AI and unless I'm missing something I don't think I can have more than one agent running because I don't have the room to spin up another AI
2026-10-24 16:28:55
0
Vyvvan Basterd :
correct. I set up an SQL database and use Claude MD at the global level to make sure it had memory and never had to relearn every command and session.
2026-10-24 12:58:26
0
Barry Carey :
pretty much what I've landed on. I keep to persistent sessions, a dispatcher and reviewer. the dispatcher spawns labeled backlog items in fresh Claude sessions. this session get a skill that defines the dev workflow. They interact with the reviewer once CI passes. when their PR is merged the dispatcher closes our the session.
2026-10-24 15:06:16
0
Metadude :
How to do this?
2026-10-24 01:29:45
1
Russ :
I just came to the realization that I need to do this😂😂😂
2026-10-23 22:21:59
1
0_kirk :
yeah when you learn tongo from using 200mil tokens a day to 20mil in context. everything opens up. 125 sub agents 3-7k context. and make them patent into a single call every code and function call the first time so you dont gave 355k token requests every single time
2026-10-23 23:07:42
4
Computer Garage LLC :
I talk about this in today videos. AI long term memory. but i only use the chatbot for coding.
2026-10-24 08:43:14
2
edenai :
Monty python reference? Expand pls
2026-10-23 23:21:21
0
Crazy Tech Dude :
Fantastic Operator skills
2026-10-23 22:38:00
0
Joey :
Infinite usage $0 money 0 bitches
2026-10-24 07:02:47
1
To see more videos from user @chiefgyk3d, please go to the Tikwm
homepage.
