How to Supercharge Your Python Development in VS Code with the March 2026 Release

Introduction

If you work with Python in Visual Studio Code, the March 2026 update to the Python extension and Pylance brings two powerful enhancements that can transform how you navigate code and speed up IntelliSense. This guide walks you through enabling and using these features step by step. Whether you're exploring unfamiliar libraries or dealing with large projects, these tools — Search Python Symbols in Installed Packages and the experimental Rust-Based Parallel Indexer — give you more control and performance. By the end, you'll know exactly how to activate them for your workflow.

What You Need

• Visual Studio Code (latest stable version recommended)
• Python extension for VS Code (version that includes March 2026 release or later)
• Pylance extension (updated to match the Python extension release)
• A Python project with an active virtual environment (venv, pipenv, conda, etc.)
• Third-party packages installed in that environment (to test symbol search)
• Optional: a larger project (100+ files) to fully appreciate the parallel indexer

Step 1: Enable Symbol Search in Installed Packages

This feature lets Workspace Symbol search (Cmd/Ctrl+T) find functions and classes inside packages from your active virtual environment. It's opt-in to avoid performance overhead on smaller projects.

Step 1.1 – Open Settings

Press Cmd+, (macOS) or Ctrl+, (Windows/Linux) to open VS Code settings.

Step 1.2 – Find the Setting

Type Include Venv In Workspace Symbols in the search bar. The setting will appear under Python › Analysis.

Step 1.3 – Enable It

Check the box next to Python › Analysis: Include Venv In Workspace Symbols.

Step 1.4 – (Optional) Tune Package Index Depth

To control how deeply Pylance searches into sub-modules of a package, look for the setting Python › Analysis: Package Index Depths. You can specify per-package depth limits (e.g., {"numpy": 2, "pandas": 1}).

Step 1.5 – Use It

Now press Cmd+T or Ctrl+T and type a symbol name from an installed package (like DataFrame or array). Results will include definitions from your environment's site-packages. You can click any result to navigate directly into the library's code.

Step 2: Activate the Rust-Based Parallel Indexer (Experimental)

This setting replaces Pylance's default indexer with a new Rust-based implementation that runs out-of-process. It promises up to 10× faster indexing on large projects, leading to snappier completions and symbol searches.

Step 2.1 – Open Settings

Again, go to settings via Cmd+, or Ctrl+,.

Step 2.2 – Search for Parallel Indexing

Type Parallel Indexing in the settings search bar.

Step 2.3 – Enable the Experimental Setting

Check the box Enable Parallel Indexing (Experimental) under Python › Analysis. Alternatively, you can add this line to your settings.json:

"python.analysis.enableParallelIndexing": true

Step 2.4 – Reload VS Code

To ensure the new indexer starts cleanly, open the Command Palette (Cmd+Shift+P / Ctrl+Shift+P) and run Developer: Reload Window. Wait a moment for the indexer to rebuild its cache.

Step 2.5 – Verify Performance

Open a large Python project and observe the speed of autocomplete, auto-imports, and workspace symbol search after the initial indexing completes. On smaller projects, you might not notice a difference — the feature is designed for scale.

Tips for Best Results

• Combine both features: Enable symbol search in packages alongside the parallel indexer to get the most out of code navigation on large codebases.
• Monitor performance: If enabling symbol search in packages slows down your workspace, reduce the Package Index Depths or disable the feature for specific heavy libraries.
• Provide feedback: Since the parallel indexer is experimental, share your experience via VS Code's feedback tool — it helps the team improve before making it default.
• Stick to active environment: The symbol search only works with packages installed in your active virtual environment. Make sure you have selected the correct interpreter (Cmd+Shift+P → Python: Select Interpreter).
• Check changelogs: For a full list of improvements, refer to the official changelogs for the Python extension and Pylance. They contain bug fixes and smaller enhancements not covered here.

Crucial Notes

• The symbol search respects py.typed markers: packages without a py.typed file only expose symbols via __init__.py or __all__. This keeps results focused and avoids clutter.
• Both settings are opt-in or experimental to maintain a solid default experience. Turning them on is safe but may use more CPU initially.
• If you disable the parallel indexer later, simply uncheck the setting and reload the window.