This repository contains the companion files for the DeepLearning.AI AI Coding Workflows course. It includes the project specifications, Go and npm dependency manifests, and custom implementer-agent configuration you'll use to build Rent a Human — a simple full-stack web app with a Go/Gin JSON API and a Vue 3 and shadcn-vue front end — throughout the course.
The simplest way to take this course is to follow the videos in order, using the same project throughout. During setup in Lesson 2, you'll create your own project and copy in the starting files from Lessons_02-08. From there, the course itself walks you through every change that appears in the later folders, so you won't need to copy anything else.
The repo is organized into three folders. Each holds the project files as they should look during that stretch of the course:
| Folder | Contents |
|---|---|
Lessons_02-08 |
The starting files: project specs, go.mod, and web/package.json |
Lessons_09 |
Adds the implementer subagent configuration created in Lesson 9 |
Lessons_11-12 |
The three-phase roadmap and the local-model implementer from Lesson 11 |
The later folders are there so you can jump into the course partway through, or check your own files against a known-good state.
A few notes:
- The
Lessons_09folder already contains the implementer file that Lesson 9 teaches you to create. If you're working from that folder while following Lesson 9 itself, delete.opencode/agents/implementer.mdfirst and let the lesson's prompt create it. If you're jumping in after Lesson 9, keep it as is. - Likewise, the roadmap in
Lessons_11-12already has the Phase 3 split that Lesson 11 has you make, so skip the roadmap-split prompt if you start from that folder. - The course was recorded on a Mac, so the local-model references in the implementer configuration and prompts use Mac values: MLX model builds. On Windows or Linux, substitute your GGUF model ID from LM Studio. The validation command,
go test . && npm --prefix web test, is the same on every platform.
All prompts used in the course are collected in prompts.md, so you can copy them rather than typing them from the videos.
| Path | Purpose |
|---|---|
prompts.md |
Every prompt used in the course, with notes on when and how to run each one. Keep this file open while you follow along |
specs/mission.md (in each folder) |
Describes the app, its purpose, and the experience it should provide |
specs/tech-stack.md (in each folder) |
Records the required technologies and implementation constraints |
specs/roadmap.md (in each folder) |
Breaks the app into phases that coding agents can plan and implement |
.opencode/agents/implementer.md |
Defines the custom implementer subagent (in Lessons_09 and Lessons_11-12 only) |
go.mod (in each folder) |
Declares the Go module and the Gin dependency |
web/package.json (in each folder) |
Lists the npm packages required by the Vue front end |
.gitignore |
Excludes build output, node_modules, local configuration, and editor files from Git |
- Go 1.24 or later
- Node.js 24 LTS or later (ships with npm)
- Git
- An IDE or editor: the course uses VS Code
- Claude Code
- OpenCode
- An OpenRouter account
- LM Studio
This project is derived from https://learn.deeplearning.ai/courses/ai-coding-workflows-from-cloud-to-local/lesson/0kjosd2k/set-up-your-environment