가입 후 초대 링크를 공유하면 동영상 재생 및 초대 보상을 받을 수 있습니다.

Michael | AI + Agents
@Michael_Fenech_
🤖 Building AI agents that do real work. 📈 3 startups, $8M raised, still shipping. 👀 Sharing everything I learn from building.
가입 August 2024
134 팔로잉 중    59.4K 팬
If you run a Grok Bot team, check whether you’ve built a team or hired the same employee five times. Run this audit to find overlapping jobs, broken handoffs and routines that fire without producing anything useful. Copy this prompt into your Chief of Staff: “Run a Grok Bot team-structure audit using the bots and records you can access. For each bot, show: 1 - The outcome it owns. 2 - What it produces. 3 - Who uses that output. 4 - When it should run. Then check: Are two bots responsible for the same outcome or decision? Is anyone actually using each output? Are routines running without useful results, or blocked by a missing dependency? Which overlaps are useful independent checks, and which are unnecessary duplication? For each issue, recommend one action: keep, clarify, narrow, merge or retire. Include the evidence and one trade-off. Finish with a one-line handoff rule showing who produces what, who receives it and who owns the next step. Keep the report compact: one table and up to three findings with concrete examples. Flag anything you cannot verify. Don’t change anything yet.” The useful part is checking what happens between bots. A research bot can produce a good report while the bot that needs it never receives it. Two bots can both do their jobs well while duplicating the same work. And a rarely used bot may still earn its place if it handles an important occasional job. Every bot needs a clear responsibility. Every handoff needs someone responsible for the next step.
더 보기