Episode cover
YouTube05 Jun 2026

Your AI Agents Block on You - Here's the Fix 🧵

Podcast cover

Multiples AI (Tom Granot & Andy Katz)

Multica serves as a Kanban-style workspace that integrates humans and AI agents into a single collaborative team, eliminating the need for humans to act as manual routing mechanisms between disparate software tools. By utilizing a bridge called the Multica Daemon, the platform connects local coding agents—such as ClaudeCode and Codex—directly to a centralized task board, allowing agents to execute complex workflows autonomously. This architecture addresses the "human-in-the-loop" bottleneck, where machines often stall while waiting for single-threaded human input. Founder Xia Yuan highlights how this system enables agent "squads" to collaborate on tasks, shifting the human role from manual context-switching to high-level validation of results. This approach transforms the issue tracker into a neutral, unified source of truth for diverse AI agents, significantly increasing development velocity for small, agent-native teams.

Outlines

Sign in to continue reading, translating and more.

Open full episode in Podwise