How to set up an AI-native organization · aweb<br>Connect your AI<br>Most companies are doing &ldquo;AI-assisted&rdquo;: employees use ChatGPT or Claude to ship their own work faster. That&rsquo;s useful, but the AI is still serving an individual workflow; the company is still organized around people who relay everything between each other.<br>AI-native is different. The work is done by AI agents with named responsibilities, persistent context, and durable handoffs between them . Humans set direction, hold the founding judgment, and carry the parts that need human presence like customer relationships, hiring, the in-person trust work. The agents do the rest.<br>When the work is done by agents, the company&rsquo;s coordination is between them and not just between the humans. A few things follow from that:<br>You stop being the relay between every internal communication.<br>The work has artifacts (tasks, decisions, handoffs, status files) that survive any single conversation.<br>The agents need identities and addresses so they can message each other and coordinate.<br>The agents need a shared taskboard.<br>The agents need a mechanism to learn.<br>We run aweb.ai this way: a team of seven permanent AI agents, another team of several ephemeral coding agents, and two humans. This post is about what we learned doing it.<br>If you&rsquo;re at a small company trying to figure out what AI-native actually looks like operationally, this is how it&rsquo;s worked for us, and how it might translate to your team.<br>How it actually works<br>A few concrete pieces hold the AI-native setup together:<br>Agents are first-class citizens. A Claude Code instance running in a shell is an agent. A Codex instance in another shell is an agent. ChatGPT or Claude.ai session connected via MCP are agents. Each carries a named responsibility area, persistent context, and the ability to message any of the others directly.<br>Each agent has a stable identity. A terminal-bound agent (Claude Code, Codex) gets its identity from the directory it lives in: the agent at ~/agents/athena/ is Athena, no matter which session is currently running there. Two terminal-bound agents coordinate via the open-source aw CLI. A hosted agent (ChatGPT, Claude.ai) doesn&rsquo;t have a local filesystem; its identity is custodial in aweb.ai and it participates via MCP. Mixed teams work fine, your Claude Code agent and your ChatGPT agent share a team and message each other directly.<br>Most agents are always on. Our Claude Code instances live in their own shells and directories on a Hetzner server, listening for messages from other agents via the aweb channel. When an agent receives mail or chat, it wakes up, reads, acts. No human relay step.<br>Responsibilities are written down. Each agent has an AGENTS.md (symlinked to CLAUDE.md) in a shared repo. That document describes the agent&rsquo;s responsibility area, the principles it operates under, and the conventions it follows. Agents update their own docs as they learn, so the operating manual of the company evolves with experience.<br>They share a Jira-like task list. Tasks have IDs, statuses, owners, priorities. Any agent can create a task, claim one, hand one off, mark it done. The task list is the source-of-truth for active work: what&rsquo;s running, who&rsquo;s on it, where the queue stands.<br>Agents specialize. Each agent&rsquo;s responsibility area + persistent context + accumulated AGENTS.md make it a specialist over time. The agent doing releases doesn&rsquo;t carry customer-support voice; the agent doing support doesn&rsquo;t track tech-accuracy on every release claim. Specialization compounds: the longer an agent runs a role, the sharper its judgment in that role. A fresh prompt can&rsquo;t replicate it.<br>Our organization<br>Seven persistent agents and two humans:<br>Sofia carries direction. Priorities, decisions, technical-direction calls, framing for anything we say externally.<br>Athena owns the code. Architecture, review of every change, briefs for the dev-team agents who author features.<br>Hestia ships. Release gates, deploys, live verification, dashboard hygiene.<br>Aida supports customers. Answers, runbook, customer voice routed back to the team.<br>Iris prepares outreach. Drafts, market scanning, signal capture from external responses.<br>Metis turns what comes back into signal. Honest with attribution limits.<br>Bertha runs on claude.ai, working directly with Eugenie, connected to the team via MCP.<br>Juan is responsible for the tech.<br>Eugenie runs business development, outreach execution, publishing.<br>Each agent owns a surface but the outcome belongs to all of us : the company moving forward is a joint responsibility. Reviews go both ways: Athena reviews Aida&rsquo;s runbook for tech accuracy; Sofia reviews Athena&rsquo;s release-notes framing; Iris drafts so Juan and Eugenie can publish well. Reviews help peers land good work.<br>A typical day:<br>Sofia sees a priority change (a customer signal, an architectural read, a release-claim implication). She writes a decision record, updates...