Telemetry-aware AI agents that act in Slack, Linear, and GitHub, without your production data leaving your cloud
Most AI observability assumes you'll ship your telemetry to a vendor's cloud so their model can reason over it. If your data is sensitive or high-volume, that's a non-starter. So we did it the other way around: the agent runs where the data already lives.
We're building groundcover, and we're spending an evening in SF showing how the new expansion to Agent Mode actually works. The short version: agents that read live production telemetry and act across your dev toolchain (Slack, Linear, GitHub) through remote MCP connectors, every action attributable to a specific user.
The main event is a live demo. We trigger a real anomaly on stage, Agent Mode ingests the telemetry, isolates the cause, and spins up an agentic workflow that opens a patch through GitHub. No recorded run, no slideware.
We'll also walk through how custom Skills map LLM-driven logic to a real infrastructure runbook, and how MCP authorization keeps every agent action tied to a user and fully revocable, all inside your own cloud under BYOC.
Food, drinks, and time to talk. The deep architecture questions, we'll connect you straight to the team that built it.
Time & Date: July 29, 2026
Location: Mindspace, 575 Market Street, 4th floor
Doors open 5:30pm
Networking: 5:30pm - 6:30pm
Speaker One: 6:30pm - 7:10pm — Matt Brown, Solution Engineer, groundcover
Speaker Two: 7:10pm - 7:30pm — Maggie Baxter, Solution Engineer, incident.io
Networking: 7:30pm - 8:00pm
About the Speakers:
Matt is a Solution Engineer at groundcover, where he works with platform and SRE teams deploying eBPF-based observability in their own cloud environments. His focus is on helping teams understand what full data ownership actually changes in practice — how they instrument, what they retain, and what it costs. He has spent nearly two decades in technical pre-sales, working alongside engineering and security teams as they evaluate and run real systems. Before groundcover, Matthew worked at Sysdig, Tromzo, Cycode, and Veracode across cloud-native runtime security, software supply chain security, and application security. He likes Kubernetes-scale problems and conversations that skip the buzzwords.
Maggie Baxter
Maggie Baxter is a Solutions Engineer at incident.io, where she helps teams design and implement thoughtful approaches to incident management.
About groundcover:
groundcover is an eBPF-based observability platform for Kubernetes and cloud-native systems — and it now goes deep on AI. Its AI Observability gives real-time, code-free visibility into LLM and agentic applications: multi-turn agents, RAG pipelines, and tool-augmented workflows, with no SDKs or instrumentation to bolt on. Because it captures at the kernel layer, it sees every interaction with providers like OpenAI and Anthropic — prompts, completions, token usage, latency, errors, and the full reasoning path. That means you can trace a hallucination back through its session context, spot where an agent misfired on a tool call, and catch quality drift — in production, without a separate stack.
Then there's Agent Mode: instead of you staring at dashboards and running manual runbooks, an AI agent picks up the signal, gathers context, forms hypotheses, and either resolves the issue or hands you a concise incident brief. And since groundcover holds more production telemetry than anyone else, that context can flow straight into your own AI agents and IDEs. All of it runs in your own cloud — no prompts, completions, or PII ever leave your environment.