One agent with twenty tools gets confused. Several small agents, each with a narrow job and a narrow toolset, do not.
This week you add sub-agents and a task tool that spawns them, parallel orchestration so independent work happens concurrently, and the design discipline of keeping each agent small and focused.
Then we assemble the capstone. By Friday of Week 12 you do not have notes. You have a tool — a complete, working AI coding agent you built from scratch, with streaming chat, a full toolset, semantic codebase search, planning, memory, a secure sandbox, tracing and multi-agent orchestration.
Something real to show, and the deep understanding to explain every line of it.
The walkthrough
Each part adds one capability to the same growing codebase, and each is a runnable checkpoint.
- 01Sub-agentsSpawn a fresh agent with its own context and a narrow toolset.
- 02The Task toolDelegation as a first-class tool the main agent can call.
- 03Parallel orchestrationIndependent work runs concurrently and results merge cleanly.
- 04Small focused agentsWhy narrower scope beats a bigger prompt.
- 05Capstone assemblyEverything from Weeks 4–12, integrated and shipped.
After this week you can
- Decompose work across specialized sub-agents
- Orchestrate parallel agent work safely
- Ship a complete, portfolio-ready AI coding agent