Keeping Knowledge Current
Andrew and Eric discuss the challenges of keeping models up to date, emphasizing the importance of version controlling all aspects to maintain performance. They draw parallels between these issues and traditional MLS problems, highlighting the complexity of managing evolving data sources in production environments.In this clip
From this podcast
MLOps.community
From MVP to Production // Day 2 Panel 2 // AI in Production Conference
Related Questions
How does version control help in machine learning, as discussed in the episode 699: The Modern Data Stack — with Harry Glaser, and in the clip Version Control for Data?
What problems do developers face when building AI applications as discussed in the episode SE Radio 610: Phillip Carter on Observability for Large Language Models and the clip Embracing New Practices?