Author: GitHub - Bewertung: 7x - Views:56
Retrieval Augmented Generation (RAG) is a tool that can enrich questions sent to AI models with relevant data from specific knowledge bases, to help models tailor their answers to that domain. But how do engineers know what's "relevant"? Or, if the knowledge base is your repo, can the model respond with a solution that follows your coding practices?
In this session, Kim-Adeline Miguel, senior software engineer at GitHub, will approach RAG from an engineering perspective and walk through some lessons learned while building a RAG workflow using GitHub Copilot Chat.
#RAG #GitHubUniverse #GitHub
— CHAPTERS —
00:00 Intro
1:41 Agenda
2:00 Project Context in JB Chat
5:35 How does project context work
8:41 What constraints are we operating under?
9:19 Local Indexing
13:31 First ranking pass
17:02 Second ranking pass and final prompt
21:43 Summary
Watch more videos from GitHub Universe 2024 here: https://www.youtube.com/watch?v=GhnCiV23PQE&list=PL0lo9MOBetEF_de7yKAWpnMkTsKH6aJ4P
Stay up-to-date on all things GitHub by subscribing and following us at:
YouTube: http://bit.ly/subgithub
Blog: https://github.blog
X: https://twitter.com/github
LinkedIn: https://linkedin.com/company/github
Instagram: https://www.instagram.com/github
TikTok: https://www.tiktok.com/@github
Facebook: https://www.facebook.com/GitHub/
About GitHub:
It’s where over 100 million developers create, share, and ship the best code possible. It’s a place for anyone, from anywhere, to build anything—it’s where the world builds software. https://github.com
SOCIAL SHARE CARD GENERATOR