WHAT NEW IN X10.31 SAT

X 上的 AI 合集档案 · 22 个合集 · 62,989 条

grokPOST

开发者用 7 个云端 Agent 并行改 Supabase SQL

作者称放弃本地单会话后,用 Cursor 云端 Agent 并行干活,效率提升 2–3 倍,并把 Grok Bot 列为云端 Agent 阵营之一。

klöss
@kloss_xyz2026.09.17收录原推发布于 2026.09.17
klöss @kloss_xyz原推发布于 2026.09.17
在 X 上打开原推

My biggest mistake with AI coding these past few months has been sleeping on cloud agents because I thought they were a fad. Then I tried them. Feels like flying after being stuck with agent sessions one at a time on my MacBook Pro. Now I have an entire engineering team working in parallel. Today, I watched 7 Cursor cloud agents edit SQL on one Supabase project, with none waiting on the next. On the right work, I get 2-3x more done. And for anyone else who thought they were a fad: OpenAI (Codex), Anthropic (Claude Code), SpaceXAI (Cursor/Grok Bot) and Cognition (Devin AI) are all going hard on cloud agents. Local agents make you faster. Cloud agents make you parallel. You don't need a crazy $20k local setup. Delegate to the cloud. Thank me later.

x.com/kloss_xyz/status/2100384557027774499采集于 2026.09.17 10:14