Powering Up Django Development With Claude Code

This is a recording from _Python Unplugged on PyTV_ – a free online Python conference by JetBrains PyCharm. Watch the talk by Sheena O'Connell, Board member of the Python Software Foundation, on powering up Django development with Claude Code. 🎤 Talk description: _Powering Up Django Development With Claude Code_ Kernighan's law, stated by the legendary programmer Brian Kernighan, observes that "everyone knows that debugging is twice as hard as writing a program in the first place. So if you’re as clever as you can be when you write it, how will you ever debug it?" The original intention of that statement was to argue for simple code. If you write fancy, clever code, then you are going to need to be extra clever when you need to fix it. Now imagine debugging someone else's fancy, clever code. Even harder, right? Despite all the hype and promise of LLM-based coding tools, the code they produce is often questionable. It's bad enough that a whole new profession has sprung up: the vibe code cleanup specialist. If Kernighan's law holds true, then what does it mean for these specialists? Vibed code is full of technical debt – puzzling architectural choices, convoluted algorithms, absurd tests (or no tests at all), badly repeated, highly coupled code... technical debt as far as the eye can see. All that being said, I kinda love working with LLMs when writing code. And there are ways to guide these tools to make them very helpful. A lot comes down to careful prompting, strategic guardrails, and healthy skepticism. I've been doing a lot of Django work lately, and Claude Code is my (mostly) trusty sidekick. In this talk, I'll be giving you a peek into how I use it and what I watch out for. This is a zero-hype talk – I'll share techniques I actually rely on and pitfalls I watch out for. 🎬 Timestamps 00:00 – Introduction 02:34 – Challenges with LLM-generated code 06:18 – Building well-structured abstractions 06:58 – Case study 08:59 – Ramping up Claude Code capabilities 10:57 – Using better models and scaffolding 12:19 – Setting up projects with LLMs 15:36 – Interactive work with AI 17:15 – Getting big things right 19:18 – Example of an idea file 21:40 – Planning and QA in development 26:22 – Conclusion 26:47 – Q&A session ▶️ Watch the full conference: https://www.youtube.com/watch?v=qKkyBhXIJJU ▶️ Explore more Python talks: https://youtube.com/playlist?list=PLCTHcU1KoD99wZ0UjDhMtxweI1JL82kcb&si=3Fb3eGClaUJYXTE5 _Python Unplugged on PyTV_ features 7+ hours of Python content across 15 talks – covering AI, web development, data science, and more. 🗣️ Hear from people behind the tools and communities shaping Python today, including Carol Willing, Paul Everitt, Sheena O’Connell, and others. #Python #PyCharm #JetBrains #PythonUnplugged #PyTV #OnlineConference #AI #Django #ClaudeCode