Separate conversational planning, codebase implementation, reusable skills, and controlled workflows.
Loading progress…
Lesson 1
AI can accelerate research synthesis, specification, implementation, testing, and documentation when the task and review standard are explicit. Learn the founder decision, practical evidence, and warning signs that matter.
Lesson 2
Prompting supplies an instruction; prompt engineering deliberately structures context, constraints, examples, tools, and evaluation for a repeatable result. Learn the founder decision, practical evidence, and warning signs that matter.
Lesson 3
Context explains the situation, constraints define permitted boundaries, and examples demonstrate the intended shape or standard of an output. Learn the founder decision, practical evidence, and warning signs that matter.
Lesson 4
Vibe-coding platforms turn conversational instructions into running software while abstracting much of the code, environment, and deployment work. Learn the founder decision, practical evidence, and warning signs that matter.
Lesson 5
Chat agents primarily reason and communicate in conversation, while coding agents inspect and change a codebase using development tools. Learn the founder decision, practical evidence, and warning signs that matter.
Lesson 6
Understand when to use a conversational assistant, a coding agent, a vibe-coding environment, or a multi-step workflow.
Lesson 7
ChatGPT is a general conversational workspace, while Codex is oriented towards carrying out software work against a repository and development environment. Learn the founder decision, practical evidence, and warning signs that matter.
Lesson 8
Claude provides conversational reasoning, while Claude Code is an agentic coding environment designed to inspect, modify, and verify software projects. Learn the founder decision, practical evidence, and warning signs that matter.
Lesson 9
An agent skill is a reusable package of instructions, resources, and workflow conventions that guides an agent through a specialised task. Learn the founder decision, practical evidence, and warning signs that matter.
Lesson 10
Skill review examines source, permissions, dependencies, instructions, outputs, and maintenance before an agent is allowed to use the package. Learn the founder decision, practical evidence, and warning signs that matter.
Lesson 11
Agent systems coordinate models, tools, memory, instructions, and workflows so software can pursue multi-step tasks with varying autonomy. Learn the founder decision, practical evidence, and warning signs that matter.
Lesson 12
An implementation prompt translates a product idea into bounded work with repository context, requirements, constraints, acceptance criteria, and verification steps. Learn the founder decision, practical evidence, and warning signs that matter.
Lesson 13
Prompt review checks whether the requested change, scope, source context, constraints, evidence, and stopping conditions are complete before execution. Learn the founder decision, practical evidence, and warning signs that matter.
Lesson 14
Testing AI-generated software verifies observable behaviour, boundaries, security, and maintainability without assuming generated code is correct or incorrect by origin. Learn the founder decision, practical evidence, and warning signs that matter.
Lesson 15
Prompt improvement after failure identifies whether the cause was missing context, an unclear decision, weak constraints, tool error, or inadequate verification. Learn the founder decision, practical evidence, and warning signs that matter.
Lesson 16
Human engineering review is required when product consequence, system uncertainty, or operational responsibility exceeds what the team can confidently verify through the AI workflow. Learn the founder decision, practical evidence, and warning signs that matter.