Running a Leaner Company With AI Agents Instead of Headcount

Frank Greeff sold his previous business for $180 million with about 400 team members. For his new company, Founders Table, the goal is different: hire fewer people, make each one as effective as possible, and fill the gaps with AI agents.

The Org Chart

The structure starts with a "Chief Agent" at the top. Think of it as a router. You send it a message through WhatsApp, Gmail, or Slack, and it figures out which AI agent department should handle the task.

Founders Table AI Organisation Chart showing the executive team sheet with Chief Agent in the center connecting to department heads

Each AI department has a "brain" that stores domain-specific knowledge. The sales department brain holds pricing and contract information. The admin brain knows scheduling and email protocols. When a human team member has a question, they ask their department's assistant, and it pulls from the relevant brain instead of making them dig through Slack threads or Google Drive.

Detailed org chart showing Sales & Admin departments with AI Head of Sales and AI Head of Admin, each branching into specialized agent roles like Outbound Sales, Inbound Reports, and Customer Topics

The founders, Jacques and Frank, sit at the top with access to query any agent across any department.

Prioritizing What to Build First

Rather than building everything at once, they are starting with the agents that make the biggest immediate impact:

CRM Enricher - When someone fills out a Google Typeform, this agent grabs public information about them and enriches the CRM record automatically.
Google Meeting Manager - Handles scheduling changes through natural language commands.
Email Manager - Routes and drafts emails based on context from the department brain.

Frank Greeff speaking to camera with text overlay: "Sold first biz for $180mil, with 400 people. Using AI agents this time"

The Operating Principle

The team follows one discipline: if a task happens more than once, evaluate whether AI can handle it. They are documenting what works and what does not as they go, planning to share weekly updates on the build process.

Key Takeaways

A central "Chief Agent" router directs tasks to specialized AI department agents
Each department has a knowledge "brain" storing domain-specific information
Start with the highest-impact agents first (CRM enrichment, scheduling, email)
The rule: if you do a task more than once, look for a way to automate it