Get this right once and stop thinking about it. The goal is not to use the fanciest tool. It's to remove friction between you and the data.

Editors and IDEs

Tool Best for Tier
VS Code (recommended default) Everyone: flexible, a huge extension ecosystem, and free Junior to Senior+
PyCharm Data engineers writing serious Python: best-in-class debugging, dbt Core integration Mid to Senior+
DataGrip / DBeaver SQL-heavy work across multiple databases. DBeaver is free; DataGrip is JetBrains-paid Junior to Senior+
Jupyter / JupyterLab Exploratory analysis and notebooks: still the default for ad-hoc work Junior to Senior+
Cursor AI-native editing, genuinely useful once you can already judge whether the AI's output is right Mid to Senior+

If you only pick one: VS Code. Add the SQL formatter, Python, Jupyter, GitLens and dbt Power User extensions, and you've covered 90% of both tracks.

Git and GitHub: non-negotiable in 2026

This isn't optional for analysts any more either (see the Skill Guide: Data Analyst (2026) doc). The minimum bar:

Junior: learn the commands above on a personal repo. See the Git & GitHub Cheatsheet for the exact commands.

Mid: comfortable with rebasing, resolving merge conflicts, and writing a PR description someone else can review from.

Senior+: sets branching conventions and review standards for a team.

Official docs: Git, GitHub

Python environment management

<aside> ⚠️

Stop installing packages globally. It will eventually break something.

</aside>

Tool Take
venv Built into Python, fine as a starting point
Poetry Popular, and handles dependency locking well
uv (recommended) Astral's fast, modern package and environment manager. Increasingly the 2026 default because it's dramatically faster and replaces pip, venv and Poetry-style locking in one tool

Starter checklist