The Programmer's Apprentice Season 2 - Advancements and Future Directions in AI-assisted Coding

Explore the advancements and future directions in AI-assisted coding in this conference talk by Erik Meijer, recorded at YOW! Australia 2023. Delve into the evolution of the programmer's apprentice concept, from its inception in 1976 to the current landscape of generative AI models. Discover how Meta has applied AI in various aspects of software engineering, including code search, recommendations, and bug fixing. Examine the potential of large language models (LLMs) to revolutionize productivity tools, transitioning from task-specific to user-specific solutions. Gain insights into the virtuous and vicious cycles of AI development, non-monotonic logic, and the power of LLM-based software. Consider the implications of automating programming tasks and the future role of developers in an AI-driven landscape.

Intro
Non-monotonic logic
Virtuous cycle
Vicious cycle
AI in software engineering at Facebook
Code searching using natural language
Code recommendations
Automated bug fixing
Let's automate ourselves away & have fun doing it
LLM as scripting client
LLM-based software is very powerful, but...
Outro