One dashboard for a nine-project AI operation
Jessy Mariau
I run nine production projects with a fleet of AI agents. The hard part was never any single project — it's seeing the whole operation at once: who did what, what's running right now, and what's quietly stuck.
The operation as one live graph
The operation has a cockpit
A realtime dashboard on Next.js + Supabase puts every project, agent, scheduled automation and memory bucket on one surface — with a web-push notification when something actually needs me. The org chart isn't just a diagram — it's the structure the agents run from; reshape it and the next run follows.
One agent per surface, at a glance
What's actually running
Live at dionisos-graph.vercel.app: eleven automated cron families, an eleven-bot Telegram fleet, and a memory graph with 200k+ relations — rendered in realtime, not a screenshot of a plan.
This is the system that runs my own operation — the clearest proof of multi-agent orchestration at scale I can show. (Agent names in the overview are generalised for privacy.)
Like this project
Posted Jul 12, 2026
Nine businesses, one control room. Every cron, agent, task and alert across the estate reporting into a single live dashboard — so the question "what is actually running right now?" has one answer instead of nine.
