Agents in the Sandboxes — LangChain Community London × Focused.io

An evening meetup from LangChain Community London and Focused.io exploring how to design, control and safely operate agentic AI systems in production.

The event examines the engineering challenges of running AI agents in production, covering sandboxing, tool permissions, prompt injection, trust boundaries, human-in-the-loop approval, observability and guardrails. The programme includes doors and networking from 6:30pm, a welcome, three talks on agent failure scenarios, building sandboxes and moving from demo to production, followed by a community Q&A and closing networking. It is aimed at engineers, architects, security practitioners, product builders and anyone interested in agentic AI.