Keynote: The past, present, and future of AI for application developers - Steve Sanderson

Steve Sanderson explores the history of AI from the 1960s to modern LLMs, including a live demo of a GPT-2 implementation in C.

This talk was recorded at NDC London in London, England. #ndclondon #ndcconferences #developer #softwaredeveloper Attend the next NDC conference near you: ndcconferences.com ndclondon.com/ Subscribe to our YouTube channel and learn every day: / @NDC Follow our Social Media! facebook.com/ndcconferences twitter.com/NDC_Conferences instagram.com/ndc_conferences/ #ai #ui So we all know AI is changing the software industry right now. Whether you build backend systems, web or native UIs, or embedded devices, you keep hearing it: the next generation of users will simply expect your software to carry out their tasks intelligently. Let's start with how we got here. I'll demo AI systems from the 1960s primordial soup up to the emergence of large language models (LLMs). We'll see a GPT-2 implementation in C code, and train it from scratch on nothing but NDC talks (what will it say??). We'll then build up to modern chat/assistant/agent systems. The point of this is to give you a deep intuition for the capabilities and limitations of LLM-based systems. This sets the scene for present-day app development. We'll run through practical examples of intelligent features you could add to your products today, coding up features like automatic data cleaning, sentiment scoring, anomaly detection, and the use of vision. Finally we'll consider what this means for the future: will the concept of UIs eventually be obsolete, or can it be reinvented to be much more fluid and intelligent? Let's see what we can do...