Async agents that actually work: watchlists, autonomy levels, and context
menhguin · x · 2026-11-10
Developer menhguin shares a hands-on playbook for making async agents genuinely useful, arguing that since users won't pay much more for frontier models, heavy users already prompt all day, and multiagent hasn't been cracked, async agents are the most viable path to 10x token spend.
The three-part method:
- Watch lists (most important): build triggers for follow-ups (e.g., a PR approval or a reply that blocks you), weighted across auto-trigger, daily, weekly, and monthly checks wired into existing crons; accumulate the list organically over a week.
- Scoped autonomy: assign L1-L5 autonomy levels to tasks, and instruct the agent to prioritize end-to-end-completable items or updates that would substantially change your decision rather than broad summaries.
- Context plumbing: hook the agent into your actual work context.
More from coding & agent
- Skills vs linked docs: where should AI agents' knowledge live? — IllustriousFault6218 · 2026-11-10
- Fable 5.2 models a Bugatti Tourbillon in Blender from a single prompt in 2.5 hours — socialwithaayan · 2026-11-10
- Pake: Rust-based tool packs any webpage into a lightweight desktop app, 539K downloads — Aiden_Tech_Ai · 2026-11-10
- Frontier3DBench: 145 tasks test whether coding agents can replace 3D vision pipelines — Aiden_Tech_Ai · 2026-11-10
- LASIK open-sourced: strips anti-AI obfuscation, cutting 21M tokens to 17k — morphh7 · 2026-11-10
- Human-in-the-loop is easy; keeping the human attentive is hard — 4 metrics to test if your AI approvals mean anything — alex_verem · 2026-11-10