
Map your AI stack for scale
You'll have a one-page diagram showing how your current AI tools and workflows handle growth — and where bottlenecks appear at 10x token volume.
What you'll make
AI Stack Scale Map
- Current State: your tools and their connections
- Token Flows: volume and direction of API calls
- 10x Bottlenecks: where each component fails first
- Placement Decisions: what moves edge vs server vs hybrid
Upload a photo or screenshot of your hand-drawn or digital diagram showing the system architecture and bottleneck annotations
The 4 parts
Each part ends with a check you can see.
Audit Your Current AI Tools
Map every AI tool you actually use and measure how much you're hitting each one.
Done when
You have a complete list of AI tools with actual usage numbers — API calls per week, tokens consumed, where the processing happens.
Trace Your Token Economics
Calculate the actual cost and volume of tokens flowing through your workflows.
Done when
You have cost-per-workflow numbers and can predict what 10x volume costs for each tool in your stack.
Test Real-World Response Times
Measure actual latency for your AI workflows and find where user experience breaks.
Done when
You have measured response times for each workflow and identified specific latency thresholds where your use cases become unusable.
Draw Your Scale-Ready Stack
Create the visual diagram showing current state, bottlenecks, and architectural decisions for scale.
Done when
You have a complete one-page diagram showing tools, connections, token flows, failure points, and placement decisions for 10x scale.
Capstone: you put together your AI Stack Scale Map.
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
- Produce a one-page system diagram showing your current AI tools, their token throughput, and specific bottlenecks at 10x scale
- Identify which parts of your AI stack would break first under heavy load and map alternatives
- Have concrete cost estimates and architectural decisions for scaling each AI workflow to production volume
- Know where to place processing (edge vs server vs hybrid) for each tool based on latency and cost constraints
Who it's for
- Teams already using multiple AI tools who need to think about scaling before hitting walls
- Technical leaders planning AI infrastructure beyond prototype stage
- Anyone running AI workflows who's wondered what happens when usage goes from dozens to thousands of requests
What it won't cover
- Building agents from scratch — this is about mapping what you already use
- Model training or fine-tuning infrastructure
- Enterprise procurement processes or vendor negotiations
- Generic AI strategy frameworks — this is about your specific stack
Skills you'll practise
- Architecture bottleneck analysis
- Cost modeling under load
- Edge-vs-server placement decisions
- Infrastructure capacity planning