Engineering in a Post-AI World explores what Applied AI Engineering looks like in this new era where state-of-the-art models are at our fingertips, ready to be adapted and integrated into real-world systems. Rather than being bogged down by every detail of gradient descent or traditional data science principles, this talk focuses on understanding the core building blocks such as tokenizers, embedders, multi-head attention, encoders, and decoders; learning how to confidently wield them to create meaningful applications.
Drawing from recent experience teaching a full AI course to seasoned developers, this session highlights what truly matters when preparing engineers to think critically about AI integration. Attendees will come away with practical insights into cutting through the noise, building reliable agents, and recognizing that data collection and system design are not new problems—what’s new is learning to align them with what modern AI requires to be useful. The talk also emphasizes the importance of observability, human-in-the-loop practices, transparency, and evaluation, which have become essential disciplines in themselves in today’s AI landscape.
