News
The dashboard is dead: agent-driven security on the platforms you’re building
The dashboard isn’t dead. It’s the end point.
“The dashboard is dead,” claims Shantanu Gattani, SVP Product at Sysdig, during Edgecase 2026 in Hilversum. Well, not exactly. The dashboard is no longer the destination. And that shift is being driven by the arrival of coding agents. Judging by the raised hands in the audience, these agents are still far from mainstream. Yet they already have the potential to resolve security incidents dramatically faster than traditional approaches. Instead of tickets bouncing between teams and analysts spending hours gathering context, agents can investigate and remediate issues in seconds.
Security is one of the dominant themes at Edgecase 2026, and according to Gattani, the biggest changes are not happening at the technology layer. They are happening in culture, workflows, and architecture. They have to. Attackers increasingly operate with automation, while many security teams still rely on a process centered around dashboards, triage queues, ticketing systems, and manual investigations. Closing that gap requires a fundamentally different approach.
The future is headless
According to Gattani, the future of security is "headless security", powered by coding agents. These agents can move through a security workflow in anything from a few seconds to ten minutes, completing tasks that would otherwise require multiple teams and handoffs. This also changes the meaning of the user interface itself.
Instead of working through screens filled with tables, dashboards, and buttons, security professionals interact directly with agents, instructing them through natural language. The role of the interface shifts from clicking to conversing. Gattani believes that makes security operations not only faster, but fundamentally more powerful. "An interface you can talk to is a hundred times more powerful than an interface you have to click," he says.
Edgecase 2026: Gotta read 'em all
All the speakers, sessions, and insights from Edgecase 2026 in one place.
Agents are teammates
Coding agents may be central to the future of security, but that raises an obvious question: how do you ensure they operate safely?
Gattani describes them as "hyper-intelligent five-year-olds". Capable, fast, and resourceful, but still requiring guardrails. The rise of autonomous agents does not eliminate the need for human oversight. Quite the opposite. As agents gain more autonomy, organizations need stronger control mechanisms. Every action must be traceable. Teams must be able to reconstruct what an agent did, which decisions it made, and why it took specific actions. "An agent that interprets an instruction independently may go further than the user intended," Gattani warns.
That means actions should be auditable and reversible. Just as importantly, organizations must start treating agents as members of the team. "Build tools that agents can communicate with."
A new role for humans
Far from becoming less important, people become more important in this model. Their role simply shifts. Humans move away from repetitive investigation and coordination tasks and towards supervision, decision-making, and governance. The dashboard does not disappear, but it is no longer where the process begins.
It becomes where the process ends. The dashboard remains the place where teams verify outcomes, review decisions, and maintain control. The heavy lifting happens elsewhere, carried out by agents working across systems at machine speed. That requires a different operating model.
And above all, it requires trust.
