The Evolution of MLOps
Talkat Cloud Native + Kubernetes AI Day NA 2024 · on Tuesday, 2024-11-12· at 10:25· Salt Palace | Level 1 | Grand Ballroom A·
summary
GMI Cloud describes itself as a GPU Cloud for AI workloads and claims improving GPU reliability compared with recent Llama 3 reports while expecting rapid inference growth. Presents concrete transitions: DevOps -> MLOps (CI/CD, Docker, Kubernetes, Terraform) -> LLMOps with continuous pre-train and fine‑tuning. Mentions tools Kubeflow, MLflow, TFX, DeepSpeed, and operational demands of hundreds or thousands of nodes for training and inference plus monitoring for model drift and safety.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 0:10About GMI Cloud
- 1:01Logistics
- 1:19History of DevOps
- 1:59MLOps
- 2:48LMUPS
- 3:42MLOps Benefits
- 4:33Key Advancements
related talks
20mBringing CI/CD Practices to Machine Learning with MLOps
14mWhy Your DevOps Skills Are Your MLOps
53mExperience with an end-to-end MLOps pipeline: Practical insights and 3D architectural perspectives
41mTHE STATE OF MLOPS - MACHINE LEARNING IN PRODUCTION AT ENTERPRISE SCALE
30mMLOps for Continuous Learning
36mIs the MLOps tooling the same as the DevOps tooling?
26mAre you ready for MLOps? 🫵
33mBuilding & Managing MLOps Pipelines in the Cloud