2026-03-22 –, Teresa Yuchengco Auditorium (Main Hall)
Air is a Python web framework built from scratch for AI-assisted development. We'll show how designing for LLM legibility makes you a better coder whether or not you use AI. This includes explicit types, clear contracts, documentation that reads like a book. The goal is a framework that works astoundingly well with commercial or open source LLMs, or even the introspection features of your LLM. While Air is being architected by the speakers of this talk, a lot of the Air codebase is collaboratively being developed at sprints in the Python Asia community, starting in Davao, Baguio, and Manila but soon growing to Singapore and beyond.
"Most Python web frameworks were designed before AI-assisted development existed. Their patterns, conventions, and documentation assume a human reader with years of context. LLMs struggle with them, even with all their documentation already ingested into the LLM catalogue. Air is different.
Built from scratch for developers who work alongside AI, Air prioritizes what makes code legible to both humans and machines: explicit types, clear contracts, disciplined testing, code examples in the docstrings for every callable that have their own tests, and documentation that reads like a book. The same practices that help an LLM understand your code help your future self, your teammates, and anyone who inherits your project. Designing for AI makes you a better coder.
These principles mean that whether or not you are using expensive hosted LLMs, open source LLMs running on your laptop, or just the introspection features of your IDE, your experience with AIr will be unusually positive.
We are here running sprints in the Philippines and soon throughout Asia, with the rest of the world to follow. Django started as a web framework built for a newspaper in the US, and its roots are clear. We want Air to be a web framework with global DNA starting with clear Filipino and Asian influence. That way we can be sure the future of the web here is free, open source, and designed for our needs here in Asia from the start."
Daniel believes great code should be accessible to everyone. His Two Scoops of Django series became the definitive guide for tens of thousands of developers worldwide, and he's contributed code to hundreds of open source packages.
From organizing the first PyCon Philippines to building django-crispy-forms, he's spent decades making Python development better for everyone. His work at NASA, Kraken Technologies, and AI research labs taught him that robust, maintainable code isn't just nice-to-have—it's essential.
When not coding, he's watching documentaries and Mark Rober videos with his daughter Uma, feeding their curiosity about how things work.
Audrey sees repetitive developer work and builds tools to eliminate it. Her project templating tool Cookiecutter saves developers countless hours—downloaded over 5 million times monthly with 24k GitHub stars. She is best known for co-authoring Two Scoops of Django, and is also a prolific fantasy and sci-fi fiction writer. She is a core developer of Air, the new AI-first Python web framework. As PyLadies' first president and cofounder in 2011, she built the organization from scratch—handcrafting the first website to the logo to the chapter kit that helped launch a global movement. MIT-trained engineer, Two Scoops co-author, and devoted mother to Uma, who keeps her grounded between building the next generation of developer tools.