Set Up a Clean Python Environment for AI

Why isolate

AI libraries have heavy, conflicting dependencies. A per-project virtual environment keeps them from clashing and makes your setup reproducible.

The modern way

# using uv (fast) — or swap for python -m venv
uv venv
source .venv/bin/activate         # Windows: .venv\Scripts\activate
uv pip install torch transformers
uv pip freeze > requirements.txt  # lock your versions

Habits that save hours

• One environment per project, never global installs.
• Commit requirements.txt (or pyproject.toml) so others can reproduce it.
• Pin versions for anything you'll deploy.