The new Grafana Cloud IRM: One picture for humans and agents

Your next incident won't be resolved by a human or an agent. It'll be both. But right now they talk past each other; agents can’t see what humans know, and agents don't see what other agents know. Everyone operates on their own slice of the picture.  

In this session, Senior Engineering Manager Mimi McNair and Staff Product Manager Michael Wettick unveil the new Grafana Cloud Incident Response and Management (IRM) solution, which provides a shared record that any human and any agent — no matter who built it — can read and write. IRM automatically captures what's being asked, what's suspected, what's known, and what's being done, while keeping one rule: Anyone can surface a finding, but only humans decide what findings to trust and how to act on them. Underneath, Grafana Cloud's knowledge graph maps your services, SLOs, and dependencies to one another, so a cascading failure becomes one incident instead of an alert storm. Incidents start with context, postmortems write themselves, and every incident makes the next one faster to resolve.