No filler days. Every day builds on the last one, inside its module, toward something you actually ship by day 50.
Meet Codex: what it's good at that Claude Code isn't
Meet Replit: build and host without leaving the browser
Choosing between them for a given task
Building a simple working app, not just a page
Adding a form that actually collects and stores data
Environment variables and keeping secrets out of your code
Debugging with the AI instead of guessing
Speeding up your workflow: prompts that save you time daily
Combining Week 1 and Week 2: same project, two tools
Week 2 project: ship a working app with real functionality
Meet Gemini: where it beats Claude and GPT, and where it doesn't
Google AI Studio: building with Gemini's API without writing much code
NotebookLM: turning your own documents into a briefable, queryable notebook
Turning a stack of PDFs into an audio Deep Dive you can actually listen to
Google Workspace + Gemini: drafting and summarizing inside Docs, Sheets, and Gmail
Multimodal prompting: feeding Gemini images, screenshots, and video
Long-context tricks: dropping in a whole codebase or book Gemini can reason over
Building a research assistant: NotebookLM and Gemini working the same source material
Combining this module with Week 1: research in NotebookLM, ship in Claude Code
Module project: a fully-sourced research brief, generated and narrated
Meet GitHub Copilot: autocomplete vs. agent mode, and when each one fires
Copilot Chat inside your editor: asking questions about code you didn't write
Microsoft Copilot in Word, Excel, and Outlook: where it actually saves time
Setting up an Azure account without getting lost in the console
Azure AI Foundry: deploying a model without becoming a cloud engineer
Copilot Studio: building a simple bot without writing a line of code
Connecting Copilot to your own business data, securely
Cost control: keeping an Azure bill from becoming a surprise
Combining this module with Week 2: prototype in Replit, harden and host on Azure
Module project: a working internal tool, deployed on Microsoft's stack
What an autonomous agent actually does differently
Giving Manus a real multi-step task and watching it work
Meta AI: what it's for, and where it fits next to the others
Making all five tools talk to each other instead of working in silos
Auditing your subscriptions: what you can now cancel
Research and content tasks, handed off end to end
Building a simple workflow that runs without you
Troubleshooting when an agent gets stuck
Putting all five modules together on one project
Final project: a complete build, using tools from all five modules
Café, farm, vineyard, construction crew, repair shop — if your business isn't one of the five tracks above, we'll build the course around it instead of the other way around. Tell us what you're building →