Grid by LimaCharlie: Automated Detection, Investigation, and Response - Full Demo

In this session, LimaCharlie CEO Maxime Lamothe-Brassard walks through Grid, LimaCharlie's agentic SecOps layer built on Claude Code, and shows how it solves security operations problems end-to-end, from initial setup to ongoing autonomous maintenance.
What's covered:

• What Grid is and how it's built on LimaCharlie's API-first SecOps infrastructure
• How to onboard a problem in plain language and let Grid build the detection, investigation, and response workflow
• The FDE (Forward Deployed Engineer) agent model and how Grid maintains what it builds as conditions change, without manual intervention
• Real use cases: ThreatLocker approval queue automation, Microsoft Entra OAuth consent attack detection, Google Workspace security posture, and automated detection engineering pipelines
• The AI Workbench and how to inspect, customize, and control every agent Grid deploys
• ROI tracking and cost visibility: automation rate, analyst hours freed, cost per case, broken down by model and agent
• Multi-tenant billing overview for MSSPs managing operations at scale
• Case management and how AI agents and human analysts work the same queue together

Grid runs on Claude Code under the hood, with your own API keys, so cost is transparent and fully in your control. Self-serve access is available at https://grid.limacharlie.io.

Timestamps:

1:00 What Grid is and what it's built for

1:47 Cost transparency, ROI measurement, and bring-your-own-key model

4:10 LimaCharlie as infrastructure and how it enables Grid

7:07 Live example: Microsoft Entra OAuth consent attack detection walkthrough

9:13 The Charter and FDE (Forward Deployed Engineer) model explained

11:31 How the FDE agent solves Day 2 problems autonomously

14:20 ThreatLocker approval queue automation use case

16:31 Additional use cases: Google Workspace, automated detection engineering, breach and attack simulation

19:09 Grid UI walkthrough: stories, connections, and infrastructure overview

22:00 AI Workbench: inspecting and customizing agents

24:35 Sessions view: cost tracking, token usage, and Claude Code under the hood

26:00 ROI tracking: automation rate, analyst hours freed, cost per case

29:30 Case management: human and AI working the same queue