Thanks. I had the AI issue again today with filling the context with words like its a mistake and so. xdamman (nprofile…99dx) you had something similar, right? Watch this video ;-)
My key takeaway from the talk:
Stay out of the “dumb zone” of the context window.
- Use only a portion of the maximum tokens for active reasoning, instead of stuffing everything with logs, JSON, and random history.
- Compact frequently: summarize earlier steps into a short, sharp markdown document and start new runs from that.
- Let subagents read and search the codebase in depth.
Have them return only the relevant files, functions, and insights, so the main agent can work with a small context.
Jurjen de Vries
npub1l7…0swft
2025-12-08 23:59:23 UTC
in reply to nevent1q…wysg
Author Public Key
npub1l77twp5l02jadkcjn6eeulv2j7y5vmf9tf3hhtq7h7rp0vzhgpzqz0swftPublished at
2025-12-08 23:59:23 UTCEvent JSON
{
"id": "3b2e36fc6bcd666bae97f38a158c0b5a08ae73ca1b6939c61259048c9ecfe697",
"pubkey": "ffbcb7069f7aa5d6db129eb39e7d8a9789466d255a637bac1ebf8617b0574044",
"created_at": 1765238363,
"kind": 1,
"tags": [
[
"e",
"7e82af8bac6d56970cb6c6ab3918da6e32b1a9610606f9289ec1904227ff65d9",
"wss://nos.lol",
"root",
"3878d95db7b854c3a0d3b2d6b7bf9bf28b36162be64326f5521ba71cf3b45a69"
],
[
"e",
"40da8579f1e2e316b0b5727af5abf3d79c4d280ef30e6dd4bd6d72eb205103c7",
"wss://nos.lol",
"reply",
"06639a386c9c1014217622ccbcf40908c4f1a0c33e23f8d6d68f4abf655f8f71"
],
[
"p",
"06639a386c9c1014217622ccbcf40908c4f1a0c33e23f8d6d68f4abf655f8f71",
"wss://relay.damus.io"
],
[
"p",
"340254e011abda2e82585cbfee4f91b3f07549a6c468fe009bf3ec7665a2e31b",
"wss://nos.lol"
],
[
"client",
"nosotros",
"31990:c6603b0f1ccfec625d9c08b753e4f774eaf7d1cf2769223125b5fd4da728019e:1728437063755"
]
],
"content": "Thanks. I had the AI issue again today with filling the context with words like its a mistake and so. nostr:nprofile1qyxhwumn8ghj7mn0wvhxcmmvqy28wumn8ghj7un9d3shjtnyv9kh2uewd9hsz9thwden5te0v4jx2m3wdehhxarj9ekxzmnyqyv8wumn8ghj7un9d3shjtnndehhyapwwdhkx6tpdsqzqdqz2nspr27696p9sh9lae8ervlsw4y6d3rglcqfhulvwej69ccmce99dx you had something similar, right? Watch this video ;-)\n\nMy key takeaway from the talk:\nStay out of the “dumb zone” of the context window.\n- Use only a portion of the maximum tokens for active reasoning, instead of stuffing everything with logs, JSON, and random history.\n- Compact frequently: summarize earlier steps into a short, sharp markdown document and start new runs from that.\n- Let subagents read and search the codebase in depth.\nHave them return only the relevant files, functions, and insights, so the main agent can work with a small context.",
"sig": "16e619133a936440991948d45ef9514efd470671fea464ebc0ed6c14e90425466dd8897bbb98d94ce9f19265c98471ada866ff32a56e8db28c749ffa01674283"
}