Agents need supervision while they run.

We’re building Sulcus so the people responsible for AI agents can see what they’re doing and step in when it matters.

The hard part starts when the agent runs.

Frameworks make it easy to define what an agent should do. Once it runs, it can loop, stall or reach for a command you’d rather check first.

Sulcus shows what is happening while the run is live, and works with the frameworks you already use instead of replacing them.

The people building Sulcus.

  • Milan Mraz

    Co-founder and CEO

    Focused on product, growth and bringing Sulcus to the teams building with AI agents.

  • Elariz Taghiyev

    Co-founder and CTO

    AI engineer and builder focused on agent infrastructure, runtime systems and developer tooling.

How we work.

  • Work with your framework, not instead of it.
  • Say plainly what Sulcus can see and do, and what it can’t.
  • Show real product behavior, not big claims.

Questions?

Email us at support@sulcus.dev.

Email us