This year, I took over two open-source Python projects that hadn’t been maintained for years. They used deprecated frameworks and old Python versions. Despite the projects being heavily io-driven and calling web-based APIs, they used synchronous methods. A refactor had been started, then never finished. There were branches full of uncommitted code, tests that tested nothing, irrelevant dependencies… you name it!

After making progress with getting the code into shape, I distilled what I learned into clear steps. When you leave this talk, you will know exactly what to do when you want to improve the performance, security and development ease of Python projects you maintain.

We’ll talk about how to fix all of the above and more, including: moving to async programming, adding type validation with Pydantic, overhauling tests, updating without anything exploding, and choosing the right dependencies.

Finally, I’ll give you a take-home checklist for updating an older project so doing this at home can be efficient and pain-free.