Keep Subagents to 3 or 4 for Optimal Performance
jasondeanlee · x · 2026-07-14
The author argues that in their use cases, 3 to 4 subagents are generally sufficient, as adding more rarely yields noticeable performance gains. Instead, excessive subagents can trigger memory pressure and swapping on macOS, especially when running multiple sessions simultaneously.
They also shared troubleshooting tips: after mapping the entire process tree, they cleared out numerous redundant background items, including stale dev servers, duplicate MCP servers, headless browsers, login items, VPN/network extensions, browser helpers, and various updaters. Post-cleanup and reboot, lag during prolonged multi-agent workloads was significantly reduced.
More from coding & agent
- Warp's six non-engineering teams all run on Linear and Claude Code — mon__lim · 2026-09-11
- Is inference latency becoming the biggest bottleneck for production AI agents? — Euphoric_Sea632 · 2026-09-11
- Anthropic researcher: 99% of engineers now run swarms of 300+ self-improving agents — AlishaOutridge · 2026-09-11
- Gergely Orosz: Shipping 10x PRs With AI Agents, Sites Fill With Small Regressions — ducha_aiki · 2026-09-11
- Same Echo Maze prompt, three frontier models: all passed visually but shipped the same hidden bug — eyishazyer · 2026-09-11
- Astra storyboards plus Minimax H3 per-shot generation boost video success rates — Hailuo_AI · 2026-09-11