swyx after a week with Devin Bots: multi-agent teams that own outcomes, not tasks
marvinvonhagen · x · 2026-11-18
Cognition launched Devin Bots, where a single Devin Bot leads a team of agents—from writing firmware to merging thousands of PRs a week. swyx, after trialing them for a week, breaks down why they're multi-agent teams done right:
- Always-on and proactive, but tuned for ROI
- Own outcomes rather than tasks, and can collaborate with other agents
- Manage teams of subagents and review themselves
- Reachable everywhere (call, text, Slack)
- Report via interactive HTML artifacts (devinapps) instead of long chats
swyx handed over his AEO, QA, cross-platform app porting, personal admin automation, and video editing work, moving from local agent worktree hell to cloud. He predicts this replaces /goal, "ultracode," loop engineering, agent swarms, and "AI employees"—the key insight being you don't want AI employees, you want automated teams that deliver outcomes.
More from coding & agent
- AI Engineering from Scratch: first-principles guide to the agent loop and context budgeting — ghumare64 · 2026-11-18
- Vivek Haldar: hundreds of tiny enterprise tools are now cheaper to vibe-code than to maintain — vivekhaldar · 2026-11-18
- Adaption ships as a Hermes Agent plugin for in-agent custom model training — sarahookr · 2026-11-18
- Agent security testing thread: capture evidence before deleting the dependency graph — krishnan · 2026-11-18
- Throwaway analysis environments: agents should leave reproducible evidence, no running servers — krishnan · 2026-11-18
- mathregistry Open Beta: An npm-Style Lean Math Registry with MCP Tools for AI Agents — DevinWatson · 2026-11-18