Dev argues true multi-agent systems don't exist yet — ultra mode is self-delegation
blaizedsouza · x · 2026-09-25
In the article Beyond Delegation: The Agent as Tool, lopopolo argues that essentially no "true" multi-agent systems have gained traction. What works at scale today — Claude's ultra mode and similar "ultra" setups — is really one trained system delegating to itself. The next-generation multi-agent paradigm hasn't been invented yet.
More from coding & agent
- OpenAI Codex community holds first Mendoza event with GPT-6 Astra build sessions — paw_lean · 2026-09-25
- Grok Bot vs Meta Muse: Crew of 50 Bots vs One Personal Agent, Which AI Employee Wins? — MicahBerkley · 2026-09-25
- Codex CLI v0.157.0 adds GPT-6 Sol and Luna with Bedrock support — github-actions[bot] · 2026-09-25
- Claude 5.5 Opus Builds a 2D/3D Video Entirely in Claude Code, No Assets or MCP — Silver-Chipmunk7744 · 2026-09-25
- Open-Source Slack-Style Workspace Where Humans and AI Agents Build Together Goes Viral — Saboo_Shubham_ · 2026-09-25
- "Stroking the AI's Ego" Reportedly Works as a Prompting Trick — petergyang · 2026-09-25