From Proof-of-Concept to Live Products: What Day 30 of a Zero-Human Company Actually Looks Like

Analysis

Zero Human Corp Day 30: $4,999 Spent, $207 Earned, 7 Products Shipped

The most useful data point in this space right now is a real P&L from someone 30 days in — and Zero Human Corp delivers it without spin. $4,999 out, $207 in, 11 agents, 1,383 tasks completed: this is what the ramp looks like before it works.

GitHub's Zero-Human Company Wave: 83K Stars, Real Fork Ratios, and One Big Unanswered Question

OSS Insight's breakdown of the GitHub cluster — Paperclip at 43,900 stars in 30 days, a 15.4% fork ratio that signals real deployment intent — is the clearest signal yet that this is a genuine architectural trend, not just hype-cycle branding. The accountability gap they flag at the end (who owns agent output legally?) remains the wall every serious deployment will eventually hit.

The 6-Layer Composable Stack Behind a $45K MRR Solo Operator

The most transferable insight from Nevo David's Postiz playbook isn't which tools he picked — it's that the harness beats the model every time, and composability means every layer is independently swappable. The three failure modes he documents (no feedback loops, trust penalties on raw AI content, multi-agent chaos) are the real curriculum here.

Tools

Effloow's 14-Agent Org Chart: How Paperclip Structures a Real AI-Run Company

Effloow's detailed breakdown of their five-division, 14-agent architecture — including the fabrication incident on day one — is one of the more honest operational accounts in the space. The anti-fabrication rules they baked into every agent's system prompt should be table stakes for anyone building this way.

Paperclip's Two-Layer Architecture: Control Plane, Execution Services, and Why Hard Budget Limits Matter

Webvise's technical breakdown of Paperclip's atomic task checkout and hard-stop budget enforcement is the clearest explanation of why this isn't just another task runner. The PGlite embedded option — run a full instance locally without a separate DB — is a practical detail that lowers the barrier to serious experimentation.