Course Wiki

AI-Native Developer Workflow — AI Dev Tools Zoomcamp Module 1

Tools change every month, but we keep working in much the same way. In this module, we take a vague product idea through specification and context. Then we implement it and run independent QA.

AI Dev Tools Zoomcamp › Module 1: AI-Native Developer Workflow › AI-Native Developer Workflow

Notes

AI-Native Developer Workflow

Tools change every month, but we keep working in much the same way. In this module, we take a vague product idea through specification and context. Then we implement it and run independent QA.

You learn:

• Spec-driven development
• Context engineering
• The product manager, software engineer, and QA roles
• Loop engineering
• Graph engineering

Recording: AI-Native Developer Workflow

Module summary: The companion article shows how to turn a vague idea into a useful specification, give an agent durable context, break the work into a backlog, and use separate product manager, software engineer, and QA roles to implement and verify the result. It uses weekly-feedback as the cautionary example of building without a spec, then walks through retroloop to show spec-driven, loop, and graph engineering end to end.

Read the article: AI-Native Development: Specifications, Loop and Graph Engineering

Example projects

We build the weekly-feedback idea twice to show why we need a specification:

• weekly-feedback is the working CLI that Claude Code produces from one vague sentence, with its source and tests included here.
• retroloop is the Django retrospective app produced after we specified the product and worked through a groomed GitHub backlog.

Homework

Complete the module with this homework:

• 2026 homework - Build a Django app with the AI tool of your choice. You don’t need to know Django.

Previous cohort materials

In 2025, we gave a fuller tour of the tool landscape. Those tools have changed, but we can still use the categories to understand what each kind of tool does.

• 2025 archived Module 1

Community notes

Add links to your notes above this line.

Key concepts

• Context Engineering
• Loop and Graph Engineering
• Spec-Driven Development

Related notes

• aidt-m02-build-and-ship-an-ai-assisted-full-stack-app
• aidt-m03-test-containerize-and-deploy-an-ai-assisted-app
• aidt-m04-devops-and-observability-for-ai-built-apps
• aidt-m05-configuration
• aidt-m05-module-5-coding-agent-capabilities-mcp-skills-pl
• mlz-m05-web-services-introduction-to-flask

Sources

• Video
• Lesson file