Skip to content
PDF ↓
← Field notes

FIELD NOTE · P1 · Personal projects

A virtual company staffed by 11 AI employees

A hand-built multi-agent orchestrator, and the trust boundary problem found inside it

11 Agents · 3 Divisions

In one line
I built a virtual company of 11 AI employees on a hand-built orchestrator, and saw firsthand the trust boundary problem when agents accept each other's work without checks.
Highlights
  • Role-based assignment, model priority with fallback, and cost tracking
  • Slack, Jira, and GitHub connections, plus a pixel-art office dashboard
  • Firsthand experience of agents trusting each other without checks
Tech
AI · Multi-Agent · Slack
ManagementTechnologySalesChief of StaffFinanceHRTech LeadSenior EngineerEngineerQA EngineerSecurity EngineerDB EngineerPlanning & SalesLeadContent Marketer(off by default)CCTVDirectivePlanReviewrequestThe error travels uncheckedThe next agent trusts it without checkingPrevious agentNext agentAdd a verification stepPrinciple from this project · not a built feature
A principle from this project: a verification step. It is not a built feature. Only documents that pass should move on.

Related cases

Read the full record in Control Plane ↗

© 2026 Kim JunHyun · The infrastructure of this site is managed as code, too

Choose a theme

Click a badge to switch to that theme on the same content. Dragging only swings it.

Back to the front page →