Session suggestions and requests

Developing the actual frontent that consumes the CMS content is on of the biggest challenges in getting a project live. In this talk, I'd like to show how you can generate your foundational code base and unblock yourself instantly with modern AI tooling.

Tired of using GPT only to beef up your boring old chatbot? Ever been in a situation where "This sounds like ChatGPT wrote it!" is more an insult than a compliment? Get ready to take your LLM-powered application to the next level: By cleverly combining modern prompt engineering and giving your LLM access to external online and offline knowledge bases and APIs, we can leverage recent LLM models (GPT4+, Claude 3+ and friends) to solve arbitrary, but complex problems in quite astonishing ways - including, but not limited to, content management and content operations. We'll explore the basic concepts of state of the art ReAct Agents - and if there's enough excitement, hack our own AI application in just a few dozen lines of python code in a quality continuation session (prerequisites: Basic python knowledge and an OpenAI API key, but that's it, I promise)

Through a brief history of content management, I'd like to illustrate why visual editing is essential in today's corporate CMS landscape. I would then proceed showing how true inline visual editing works in React Bricks, a visual headless CMS based on React components.