AI assistants can answer most technical questions in seconds now. So why would a developer sit through a guided workshop? Because reading an answer and building muscle memory are fundamentally different things. This talk looks at how to design interactive, hands-on developer education that stays relevant and engaging in a world where "just ask the AI" is the default. Drawing on experience building an open source training platform and creating and delivering workshops on it, the talk gets into the practical details of structuring workshop content, giving learners real environments to work in, keeping them engaged, and using AI behind the scenes to create that content faster without sacrificing the human judgement that makes it more than just AI slop.
Developer education is at an inflection point. AI coding assistants have made it trivially easy to get a snippet of code or a quick explanation, which means the bar for “worth my time” educational content has never been higher. As developer advocates, we need to rethink what we offer that AI cannot: curated learning journeys, real hands-on environments, and the confidence that comes from having built something yourself.
So how do you actually create workshop content that works? How do you write steps that teach concepts progressively without losing people along the way? What does it look like when you give every learner their own browser-based session with a terminal, editor, and live environment, spun up on demand? These are the kinds of questions that come up repeatedly after years of building and delivering workshops, and this talk digs into the practical answers.
The talk also tackles a tension that anyone in DevRel will recognise. On one hand, we’re building workshops specifically because we want developers to learn by doing rather than just asking an AI for the answer. On the other hand, AI is genuinely useful for us as content creators. It can help draft initial workshop material, suggest exercise variations, and speed up the tedious parts of content production. The trick is knowing where to lean on it and where your own experience and instructional judgement need to take over.
This talk grew out of building an open source training platform from scratch to solve the problems that make developer workshops painful, and then creating and delivering workshops on it. It is aimed at developer advocates, technical trainers, documentation writers, and anyone who creates learning content for developers. No specific technical background is required.
Graham is the author of mod_wsgi, the Apache module for hosting of Python web applications using the WSGI interface, and the author of wrapt, a decorator and monkey patching library for Python. He also has a keen interest in developer learning platforms, and is the creator of Educates, a Kubernetes native platform for delivering interactive workshop environments through your web browser.
