Lightning Talk: Building AI That Ops Teams Actually Trust
summary
Build AI that ops teams actually trust by replacing monolithic agents with modular subagents that separate context gathering, hypothesis formation, telemetry queries, and evidence evaluation. Validate not just final outcomes but full trajectory evaluation—every tool call and reasoning step—and ground tests in production scenarios (thousands of investigations per week, over 600 tests and hundreds of golden trajectories). Roll out in shadow mode so operators observe correctness, build confidence gradually, and get actionable explainability aligned to their mental model.
AI-written summary · about
related talks
09mLightning Talk: Lessons from Putting AI in Front of a Platform: Taming the Non-Dete...
29mBuilding Trust in AI Agents
30mThe Great Doubt: What Building an AI Agent Taught Us About Trust
25mWhen Your "New Teammate" Hallucinates: Observability as the Safety Net for AI Code
17mAI TechOps: Data and Infra for AI Foundation to Scale and Manage 1000s of Cohesive Agents
24mLightning Talk: Building the AIverse: Standardizing Autonomous AI Agents for a Decent...
05mLightning Talk - AI is Like Chocolate
26mBeyond Hype: Leading the Charge in Responsible AgentOps for Business Transformation