Vector Podcast: Code search, GitHub Copilot, LLM Prompting with empathy and Artifacts with John Berryman
People in every walk of life are much more effective and efficient at creating. If you really love debugging and writing tests, I’m sorry — your days might be numbered. But if you love creating, I think we are approaching a new golden age, and it is exponential: we gonna keep approaching new golden ages for a while.
I got a new angle in understanding, and even empathizing, with Large Language Models. John Berryman has spent several recent years helping develop GitHub Copilot, and today runs an LLM consultancy, Acturus Labs.
You might also know John as the co-author of the Relevant Search book, written together with Doug Turnbull (see the episode with Doug here).
He recently poured his LLM wisdom into his new book: “Prompt Engineering for LLMs: The Art and Science of Building Large Language Model–Based Applications”. Here is John’s bio:
John Berryman is the founder and principal consultant of Arcturus Labs, where he specializes in AI application development (Agency and RAG). As an early engineer on GitHub Copilot, John contributed to the development of its completions and chat functionalities, working at the forefront of AI-assisted coding tools. John is coauthor of “Prompt Engineering for LLMs” (O’Reilly).
Before his work on Copilot, John’s focus was search technology. His diverse experience includes helping to develop next-generation search system for the US Patent Office, building search and recommendations for Eventbrite, and contributing to GitHub’s code search infrastructure. John is also coauthor of “Relevant Search” (Manning), a book that distills his expertise in the field.
John’s unique background, spanning both cutting-edge AI applications and foundational search technologies, positions him at the forefront of innovation in LLM applications and information retrieval.
In this episode we dove into the topics of RAG and whether it is really just a prompt engineering technique, LLM predictability vs more traditional programming, how to implement a code assist (based on GitHub Copilot experience), what are artifacts for chat models and how future is shaped with LLMs.
John is a highly practical person, able to demo on the spot (I did not ask him to prepare ahead of time).
The quote in the beginning of this blog post about how LLMs change our developer lives has stuck with me. And I think it prepares us for another way of interacting with the world around us and with the ideas, that we craft inside ourselves. Rather than getting scared of this new development, like it may sometimes feel.
Oh, and last by not least: John has a unique sense of humor :)
Enjoy this episode:
Design by one and only Saurabh Rai.
Listen on your favorite platform:
More options here: https://rss.com/podcasts/vector-podcast/1888857/
