Project workbench

Move from browser experiments to a real project workspace.

Level 0 concepts are browser-first. Local Python, repositories, terminals, and virtual environments become builder skills gradually instead of appearing as hidden prerequisites.

Browser evidence→Open files→Run Python→Validate

Level 0 Core path — no local setup required

You can demonstrate the Level 0 exit skills with the browser Lab and an evidence report before you learn repository or terminal workflow.

1. Complete Lesson 0.12 and run its Lab unchanged.

2. Record before, after, and baseline.

3. Change only candidate_threshold = 3 to candidate_threshold = 2, run again, and explain the new mistake.

4. Write a short evidence report covering feature/label roles, train/test separation, the baseline, your debug hypothesis, the result, and one limitation.

Open the Level 0 project brief →

Five workspace words

These are tools, not assumed background knowledge.

Repository

The project folder

The repository contains the curriculum, Labs, tests, and project starter files.

Project root

The top folder

You are at the root when folders such as projects and website are visible.

Validator

A behavior check

A validator checks required project behavior. Passing it is evidence, not a replacement for your explanation.

First local run

Use the exact commands first. Explore after the known-good path works.

git clone https://github.com/yonpak/ai-from-zero-to-builder.git
cd ai-from-zero-to-builder
python --version

On some systems the Python command is python3. For the Level 0 Builder path, no third-party package is required.

python projects/starters/l00/data_detective.py
python projects/tests/l00/validate_submission.py projects/starters/l00/data_detective.py

Level 1 adds a virtual environment

A virtual environment keeps this project's packages separate from other Python projects.

macOS / Linux

python -m venv .venv
source .venv/bin/activate
python -m pip install -r projects/starters/l01/requirements.txt

Windows PowerShell

python -m venv .venv
.venv\Scripts\Activate.ps1
python -m pip install -r projects/starters/l01/requirements.txt

Debug the workspace before the model

When a command fails, find the earliest broken boundary instead of changing model code at random.

01LocationAre you at the repository root?

02PythonDoes the expected Python command work?

03EnvironmentIs the required virtual environment active?

04CodeOnly then debug the project logic.

Continue to project briefs →