By Warp Level · Warp 3
Warp 3 — Delegated & Specialized
Multiple agents, roles, and the bookkeeping to justify spend.
You stop running one agent and start running a small team of them. Different roles get different workflows, and you add the cost tracking that tells you whether the spend is worth the outcome. Target: ~10–20 tickets in flight — still hand-juggleable, but getting harder.
Roles & orchestration
Role files for multi-agent work. Consolidate the workflows and checklists you built at Warp 2 into roles. An Engineer differs from a blog writer; a Growth Engineer does outcome engineering — code and content that leads to sales.
Multi-agent orchestration via foreground subagents. Roles run as foreground subagents launched by the Chief of Agents. There is no inter-agent messaging— each subagent runs to completion and returns its report, so front-load each one's full instructions.
Checklists enforced by roles.(Strict mechanical checklists aren't implemented yet — enforcement is by convention.)
Design principles — do you have them written down? If so, commit them so every role inherits them.
Bookkeeping & cost tracking
Track cost per ticket and tabulate totals to see where tokens go — the only way to judge if spend is worth the outcome.
Don't spec-creep. There's always another feature; cost visibility is what keeps you honest about which ones are worth it.
Default processes for monitoring longer tasks while you're away.
Prioritization & the tech tree
Routines to help prioritize tickets and tasks. You still control risk and launch tasks one by one — routines only suggest at this level.
A tech tree to manage growing task dependencies and visualize the main themes you're juggling toward a functional product.
Notice the shape: Warp 3 is Warp 2's workflows, promoted into reusable roles and made measurable. You're not granting new autonomy yet — you're building the organization and the instrumentation that Warp 4 needs to run ahead of you.