Today I needed to create a graph for a book, and after faffing a little in Inkscape and seeing this was not the way, I reached to Jupyter notebooks with matplotlib.

VSCode has had good ipynb integration for some time, rendering the plots from Matplotlib and loading the kernel. One faff with matplotlib was consulting the matplotlib help a fair bit. But codepilot here is a game changer, predicting really well what I meant - it made putting together graphs demonstrating errors and integrals together really easy.

VSCode helping me produce python matplotlib graphs in a Jupyter Notebook I recommend if you have graphing needs to try this killer combo!

If you’ve not come across it yet, codepilot is an AI enabled system to predict code - synthesizing based on what you’ve already typed, even comments. It’s not out to steal your job, but will make it easier by filling in what you intend once you start typing it. Like code inference autocomplete but done really well.