ReviseAlgo Logo

Modules & Packages

pip and Package Management

Installing, managing, and distributing Python packages

Interview: Essential tool — expected knowledge for any Python developer role

Last Updated: June 12, 2026 • 8 min read

pip is the standard package installer for Python. It installs packages from the Python Package Index (PyPI) and manages dependencies. Understanding pip, virtual environments, and requirements files is essential for every Python developer.

Essential pip Commands

  • Install: pip install package — installs latest version from PyPI
  • Specific version: pip install package==2.0 or pip install "package>=2.0,<3.0"
  • Upgrade: pip install --upgrade package
  • Uninstall: pip uninstall package
  • List: pip list — shows installed packages and versions
  • Show: pip show package — detailed info about an installed package

Requirements Files

  • Generate: pip freeze > requirements.txt — lists all installed packages with exact versions
  • Install from: pip install -r requirements.txt
  • Version specifiers: ==, >=, <=, ~=, != — control version constraints

Modern Alternatives

  • pipenv: Combines pip and virtualenv with Pipfile.lock for deterministic builds
  • poetry: Full dependency management with pyproject.toml, lock files, and publishing
  • uv: Extremely fast Rust-based pip replacement — drop-in compatible
  • conda: Cross-language package manager popular in data science

Always Use Virtual Environments

Never install packages globally with pip. Always create a virtual environment first: python -m venv .venv && source .venv/bin/activate. This isolates project dependencies and prevents version conflicts between projects.

Use Cases

Setting up project environments with isolated dependencies

Managing team dependencies with requirements.txt or lock files

Publishing packages to PyPI for public or private distribution

Reproducible builds with pinned dependency versions

CI/CD pipeline setup with automated dependency installation

Common Mistakes

Installing packages globally instead of using virtual environments

Using pip freeze without a clean venv — includes unnecessary packages

Not pinning dependency versions — builds break when upstream changes

Forgetting to update pip itself — old pip versions have resolution bugs

Not using pyproject.toml for new projects — it's the modern standard