READ BY GOAL
Start with what you want to work on.
Three routes through the articles. Each builds from an idea to a practical exercise. Follow the order, or go straight to the question you’re working through.
PATH 01 · 5 ARTICLES
Designing an AI system
Choose the parts, define the task, give the agent useful context, then apply it to a changing design.
What you’ll work towardsA clearer system design and a task brief you can try with an AI model.
Start this path-
How AI systems fit together
Choose the model, knowledge, execution and integration parts for a supplier review.
-
Skills, MCP, RAG and Memory: How an AI Agent Investigates a 500 Error
Trace a publishing failure and distinguish the roles of skills, MCP, RAG and memory.
-
Giving AI a job description
Turn that design into a task contract with inputs, limits and an output schema.
-
Give AI agents a design reference
Give a coding agent concrete design references it can use during implementation.
-
What is Astra, and when would I use it?
Apply the approach to a booking migration and follow the decisions when requirements change.
PATH 02 · 6 ARTICLES
Improving business workflows
Understand the work people actually do before changing the software that supports it.
What you’ll work towardsA map of hidden rules, responsibilities and exceptions, with a small improvement to test.
Start this path-
Software should follow the business
Start with the operation and identify the rules your software needs to support.
-
The person missing from the diagram
Find the people who keep handoffs moving and make their responsibilities explicit.
-
What the spreadsheet knows that the system does not
Extract the rules and local knowledge stored in an everyday spreadsheet.
-
The exception path is the real workflow
Design what happens when evidence is missing, approval stalls or a service fails.
-
Kaizen, applied to software
Choose one small change and decide how you will tell whether it helped.
-
Jira and Confluence: From a Business Question to a Working Feature
Map how work happens with the people doing it, then connect agreed improvements to decisions, Jira work and validation.
PATH 03 · 5 ARTICLES
Making architecture decisions
Understand what software architecture means, model the system, choose useful views, then trace what happens at runtime.
What you’ll work towardsA design you can explain at several levels, from business rules to a running machine.
Start this path-
What is software architecture?
Understand software architecture: the parts of a system, their responsibilities, and the decisions that connect them.
-
Modelling the system before the code
Model an order process before choosing the implementation.
-
The C4 model: architecture at the right level
Turn the model into architecture views that answer different readers’ questions.
-
What happens when you press a computer’s power button?
Follow the running system down to its foundations: firmware, operating system and processes.
-
If AI can build the software, what is the architect for?
Connect AI-assisted implementation to business meaning, system ownership and a recoverable first release.
Each article includes a prompt, code, commands or a worksheet you can try. Reading times exclude the exercises. Browse all practical resources →