What are Agentic Loops?

June 9, 2026

AI Summary

5 min read

Ross Mike, a developer who works with AI tools daily, came on the show to explain what agentic loops actually are—and why he thinks most people should avoid them. His central argument is that the version of agentic loops being hyped by prominent AI figures burns money, makes too many assumptions, and only works well for people with effectively unlimited token budgets. For anyone building a real product, he argues, the human still needs to stay in the loop.

The Two Kinds of Loops

Ross draws a clear distinction between two ways of working with AI. The first is the "human in the loop" model: you prompt an AI agent, review the result, test it, and then prompt again. You are directing, governing, and approving each step. This is how most developers use tools like Cursor, Claude, or Codex. The second is the "agentic loop" that has been getting attention from figures like Boris and Peter at major AI labs. In that model, you write a spec document (a PRD.md or similar), fire it off to an agent, and the agent then feeds its own output back into itself, iterating without human intervention. The human only touches the system once, at the start.

Why the Hype Doesn't Fit Most Builders

Continue reading the full summary in the app — free to try.

Read Full Summary →

Free • No credit card required

Listen to Audio Summary Open in App

Never miss an episode of The Startup Ideas Podcast

Get every new episode summarized in your inbox — free, ~5 minutes to read.

No spam. Unsubscribe anytime.

What you'll learn

1 (00:35) **Guest Introduction: Professor Ross Mike** - Ross Mike is introduced to explain what agentic loops are, why they're hyped, and why he believes they are a "terrible mistake" for most builders.
2 (01:23) **The "Human-in-the-Loop" Baseline** - Ross diagrams the standard way most people use AI: a human prompts an agent, reviews the result, and iterates.
3 (03:20) **The "Agentic Loop" Hype** - Ross diagrams the alternative: a human fires off a loop once, and the agent feeds its own results back in to continue working autonomously.
4 (04:32) **The Core Critique: The "Smart Developer" Analogy** - Ross argues that agentic loops are like hiring a developer to build an entire app without consulting you.
5 (06:49) **The Token Cost Problem** - The biggest practical issue is that these loops burn massive amounts of tokens.
6 (08:03) **Slash Goal and Other Loop Tools** - Ross explains that tools like "Slash Goal" are all the same concept: a single prompt to build an entire thing without stopping.
7 (09:58) **When Loops Work: Experimentation** - Ross shares a personal example of using a loop to build an "Among Us simulator" for AI models.

+ Full timestamped outline available in the app

Show Notes

S/o Coderabbit for sponsoring today’s vid: https://startup-ideas-pod.link/code-rabbit

On this episode I sit down with Professor Ras Mic to break down agentic loops. We define what a loop is, explain why well-known builders like Boris and Peter swear by them, and stay honest about who they truly serve. Mic argues that human-in-the-loop remains the strongest setup today, and he walks through the one loop he runs every day for code review using Cursor, GitHub, and Greptile. By the end you will know when a loop earns its place and when your own hand belongs on the wheel.

Timestamps

00:00 – Intro

01:23 – What is a Loop

07:59 – /goal Explained

11:32 – The Slop Machine

12:42 – Code Review as a use case for Agentic Loop

18:19 – Honest Take for Builders

20:42 – The Future of Loops

21:50 – Closing Thoughts

Key Points

A loop fires once from a human, then the agent generates, reviews its own result, and feeds it back to keep building.
Human-in-the-loop keeps you directing, governing, and approving each step while the agent builds.
Wide-open loops make heavy assumptions and burn serious tokens; Michael cites Peter's tweet about $1.3 million worth of tokens in one month.
Reserve slash goal and similar loops for the $200/month plan, since the $20 and $100 tiers burn through fast.
Loops shine in confined, fixed-feedback work: code review, SEO pages, and other binary tasks.
Mic’s daily win is a closed code-review loop with Cursor, GitHub, and Greptile that chases a 5/5 score.

The #1 tool to find startup ideas/trends - https://www.ideabrowser.com

LCA helps Fortune 500s and fast-growing startups build their future - from Warner Music to Fortnite to Dropbox. We turn 'what if' into reality with AI, apps, and next-gen products https://latecheckout.agency/

The Vibe Marketer - Resources for people into vibe marketing/marketing with AI: https://www.thevibemarketer.co