
Design an AI agent role that mirrors a specific junior developer on your team
You'll have a defined AI workflow (with prompts, guardrails, and handoff rules) that handles one concrete task category your junior dev currently owns, freeing them for higher-judgment work.
What you'll make
AI Agent Workflow Design
- Task Definition & Scope: what to include
- AI Agent Instructions & Guardrails: what to include
- Human Review Checkpoints: what to include
- Handoff Protocol: what to include
- Team Integration Plan: what to include
Document your prompts, review criteria, and handoff rules as a runnable workflow your team could actually implement
The 4 parts
Each part ends with a check you can see.
Map Current Delegation Patterns
Inventory what you currently delegate to your junior developer and how you review their work
Done when
You have a prioritized list of 3-5 task categories with current delegation and review patterns documented
Test AI Agent on One Category
Design AI instructions for your highest-priority task category and test them against real examples
Done when
You have tested AI agent instructions that handle one task category, with evidence of what works and what requires human judgment
Design Review Checkpoints
Create specific review criteria and handoff rules that maintain team understanding when AI does the initial work
Done when
You have documented review checkpoints with specific criteria for accepting, revising, or rejecting AI agent output
Document Complete Workflow
Combine your AI agent instructions, review checkpoints, and handoff rules into a runnable workflow
Done when
You have a documented workflow that another team lead could implement, covering task scope, AI instructions, review criteria, and team integration
Capstone: you put together your AI Agent Workflow Design.
Where this helps
An elective: a single lesson that sits outside rtwo's job certificates and doesn't count toward one.
About this lesson
What you'll be able to do
- Design and test one AI workflow that handles a specific task your junior dev currently owns
- Build review checkpoints that maintain team understanding even with AI-generated work
- Judge when to expand vs constrain AI agent scope based on real team dynamics
Who it's for
- Engineering managers with junior developers they want to upskill
- Senior developers who delegate regularly and want to experiment with AI assistance
- Team leads managing code review processes with mixed AI/human output
What it won't cover
- Multi-agent orchestration frameworks
- Training custom AI models
- Wholesale replacement of human developers
- Enterprise AI deployment infrastructure
Skills you'll practise
- Workflow decomposition
- AI-human handoff design
- Review process optimization
- Team coordination under AI assistance