Explore the intricacies of MLOps and ML Orchestration in this 50-minute podcast episode featuring Ketan Umare, CEO of Union.AI. Delve into the relationship between Union and Flyte, emphasizing community-driven development and the challenges of balancing feature requests with security considerations. Learn about the importance of real-time data and secure data handling in orchestrating machine learning models. Discover how the Flyte community's support for newcomers contributes to democratizing machine learning. Gain insights on recent changes in orchestration, the role of community in Flyte's development, and the distinctions between Union and Flyte. Explore high-value features of Union AI, the building of LLM infrastructure, and the evolving landscape of traditional ML and prompt engineering.
MLOps vs ML Orchestration - Exploring Challenges and Community-Driven Development
Overview
Syllabus
[] Ketan's preferred coffee
[] Takeaways
[] Please like, share, and subscribe to our MLOps channels!
[] Shout out to Ketan and UnionAI for sponsoring this episode!
[] Orchestration recent changes
[] Community with Flyte
[] ML orchestration
[] 50/50 is generous
[] Real-time ML
[] Over engineering without benefits
[] Balancing everything
[] Union verse Flyte
[] High value features of Union AI at the back of Flyte
[] Building LLM infrastructure
[] Traditional ML is the whole prompting
[] LLMs to evaluating prompts
[] Wrap up
Taught by
MLOps.community
Related Courses
-
The Role of Infrastructure in ML Leveraging Open Source - MLOps Podcast #197
-
Enterprises Using MLOps - The Changing LLM Landscape and MLOps Pipelines
-
MLOps at the Crossroads - Challenges and Future Directions
-
How to Systematically Test and Evaluate LLM Apps - MLOps Podcast
-
Building an ML Platform: Insights, Community, and Advocacy - MLOps Coffee Sessions
-
Unpacking Three Types of Feature Stores - MLOps Podcast
