Micah Lyle
I'm a software architect and engineer, and co-run a design and software agency and consultancy called Elyon Technologies. Before that, I spent several years as a solo freelancer, working on a number of different projects and systems. Almost everything I did I built in Django (from scratch or on top of something existing), and I strongly focused on a niche specifically working on Django apps both through UpWork and also word of mouth referrals. I've gotten to build and architect software systems with Django across a number of different verticals which has provided both narrow/focused perspectives in the Django space but also helped me grow opinions on how codebases should be structured in the long run.
Session
Python 3.13 enabled experimental support for free threading, which disables the GIL, and allows multiple threads to execute in parallel without getting blocked by the GIL. This has huge promise and potential with Django for both performance and efficiency, because traditionally Django is run with gunicorn (or uvicorn in the asynchronous case), which will do a process per-worker. The ability to be able to use threads instead of processes and still have them execute in parallel would indeed be a considerable boost to performance and potentially memory efficiency as well, but there are a number of hurdles that will have to be overcome before this becomes a reality. This talk will explore those hurdles, highlight Django internals that may be impacted by free threading, and do a live demo of free threading performance and memory efficiency with Django.