Sarah Simionescu argues that agents can change how people interact with dashboards, using her own experience of accessing Datadog without opening its interface.
She describes how MCPModel Context Protocol is a standard way for AI applications to connect to external tools and sources of context. connects agents to services while leaving challenges around memoryAI agent memory is stored information that an agent can retrieve and use across steps, sessions, or changing contexts., tool selectionAI tool calling lets a model request a structured operation from software instead of only returning natural-language text., contextContext engineering designs how relevant instructions, evidence, state and tool information are assembled for a model's task. and cross-app coordinationAgent orchestration coordinates AI agents, tools, people, tasks, state, and control flow so a larger workflow reaches a verified outcome.. In a mockup she shows a bug investigation using Composio’s tool discovery and planning.
A second example combines analytics across PostHog and Metabase without loading all results into model context. Simionescu presents early company comparisons and advocates designing applications for agent users.
Watch on YouTube
