Explore the integration of Large Language Models and foundation models within existing software ecosystems using open-source Python in this 27-minute presentation by Hugo Bowne-Anderson, PhD. Delve into the balance between innovation and established practices in the machine learning stack, focusing on practical applications in text-to-image and text-to-speech technologies like Stable Diffusion and Whisper. Understand the importance of workflow orchestration in combining Generative AI with classical Machine Learning for robust, production-ready systems. Learn about traditional software, building blocks of ML systems, the ML infrastructure stack, and how Generative AI is changing data workflows. Discover techniques to make Generative AI valuable and get insights on implementing these concepts at home. The presentation concludes with a Q&A session, providing additional clarification on orchestrating Generative AI workflows to deliver business value.
Orchestrating Generative AI Workflows to Deliver Business Value
Overview
Syllabus
- Introductions
- Traditional Software
- Building Blocks of ML Systems
- The ML Infrastructure Stack
- How is Gen AI Changing Data Workflows?
- Towards Generative AI Systems
- How do these techniques make Gen AI valuable?
- How can I do this at home?
- Q&A
Taught by
Open Data Science
