
Compose a cost-optimization brief for leadership
You'll have a 1-2 page written analysis showing how model choice and agent persistence affect your team's monthly AI spending—with a recommendation for where to switch models or workflows.
What you'll make
AI Spend Optimization Brief
- Current State: your team's monthly AI costs broken down by tool and workflow
- Cost Drivers: where persistence vs single-shot creates the biggest spend differences
- Migration Plan: 3-month roadmap for switching tools and workflows
- Risk Mitigation: what could go wrong and how to catch it early
Write as a formal brief with specific dollar amounts and timeline commitments
The 3 parts
Each part ends with a check you can see.
Map Your Current AI Spend
Audit your team's last 30 days of AI tool usage to find where money goes and why
Done when
A breakdown showing monthly spend per tool (GPT-4, Claude, Codex, etc.) with at least 3 workflows consuming >$100/month identified by cost driver
Analyze Workflow Cost Patterns
Use AI to model how different task delegation patterns affect your monthly spend across tools
Done when
Cost pattern analysis showing how context handoffs vs single-agent approaches would change spend for your 3 highest-cost workflows
Design Your Migration Roadmap
Build a 90-day plan showing which workflows to migrate between AI tools based on your cost analysis and risk tolerance
Done when
A month-by-month migration plan with specific workflows, expected cost changes, and rollback criteria for at least 3 workflow changes
Capstone: you put together your AI Spend Optimization Brief.
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 cost-conscious workflows that delegate tasks between multiple AI agents without unnecessary usage overlap
- Document the monthly spend impact of switching between GPT-4, Claude, and specialized tools like Codex for different workstreams
- Build a 90-day migration plan showing which workflows should use persistent agents vs single-shot calls
Who it's for
- Engineering managers tracking AI tool costs across teams
- Operations leaders evaluating multi-agent tooling investments
- Technical leads designing workflows that span multiple AI systems
What it won't cover
- Setting up API billing dashboards (assume you have access to usage data)
- Negotiating enterprise contracts with AI vendors
- Building custom agent orchestration infrastructure from scratch
Skills you'll practise
- Multi-agent workflow design
- AI cost modeling
- Persistent vs ephemeral task delegation
- Cross-vendor toolchain analysis