If you're trying to run five Claude Code sessions in parallel, reaching for git worktrees you've found the right problem. But every team that pushes past this point hits the same wall: agents need isolated, connected, governed environments that localhost can't provide.
On-demand
Coming soon
What this webinar covers:
Stripe's Minions merges over a thousand agent-authored PRs per week. Ramp's background agent accounts for 57% of their merged PRs. The insight both share is that they invested heavily in standardizing their development environments long before agents. As Johannes Landgraf, CEO of Ona, wrote in The Last Year of Localhost, that infrastructure is now a hard prerequisite to enabling background agents within an organization across the SDLC.
This session covers the primitives you need for background agents at scale: sandboxed execution, governance, context and connectivity, and orchestration for agent fleets. We'll walk through the limitations of local agents, why standardizing your development environment is the first step, and how these primitives compose into a self-driving codebase.
Key takeaways:
Why worktrees and running agents locally are a dead end
The infrastructure primitives for background agents
What infra Stripe and Ramp needed to unlock background agents
This webinar is for you if:
You're an engineering leader who has adopted code assistants but hasn't seen organizational throughput improve, or a platform engineer tasked with enabling background agents and discovering that localhost and worktrees won't scale.
