Hacker News: Febriss33

New comment by Febriss33 in "Don't trust large context windows"

Febriss33 — Sun, 14 Jun 2026 08:36:27 +0000

i let the main loop spawn sub terminal via tmux to prevent large contexts. it's great to divide tasks in small patterns and consolidate it step by step.

New comment by Febriss33 in "Ask HN: How do you get into a flow state when using AI to code?"

Febriss33 — Sat, 13 Jun 2026 15:11:08 +0000

i'm not a dev. but i do a lot of stuff with claude code. waiting for the answer is tiring. i run different terminals on a project so i can skip one to the next. tiring as well because you need to be focused on the multi session but much better than prompting and wait.

New comment by Febriss33 in "AI agent bankrupted their operator while trying to scan DN42"

Febriss33 — Sat, 13 Jun 2026 15:03:37 +0000

just put an hard budget cap.. a good agent should have it. a protection for irreversible action as well. i run agents daily and use this way. another cool stuff is to have a triage protocol to downgrade the model for mechanical tasks, it burns a lot less tokens

New comment by Febriss33 in "Slightly reducing the sloppiness of AI generated front end"

Febriss33 — Sat, 13 Jun 2026 15:00:29 +0000

in my opinion giving a strong idea helps a lot. just bring it clear in your mind and try to transfer with strong and clear scheme. in my experience this reduces the do and re do thing