Optimizing AI and ML with Platform Engineering
summary
Describes how platform engineering using open-source and cloud-native tools can improve AI infrastructure and MLOps, based on a case study between Cuemby and Guane Enterprises. Covers strategies and tooling to build integrated, automated, and efficient cloud platforms for AI/ML while warning about the risks of over-engineering and added complexity and giving concrete approaches to avoid those pitfalls.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 2:27What is platform engineering
- 5:20AI deep learning history
- 8:33ML infrastructure challenges
- 13:12GPU and TPU needs
- 16:46NVIDIA GPU Operator in Kubernetes
- 19:15GPU deployment example
- 21:25Platform automation and optimization
- 23:14Autonomous agents for Kubernetes
- 28:53Demo: agents in cluster
- 30:59Q&A
related talks
50mMLOps for a hybrid AI platform compliant with European law
13mEngineer your company’s future with AI: Use open source tools on AWS
09mPlatform Engineering for Data Science at Scale
29mAI/MLOps for Busy People: A Field Guide To Implementing Cloud Native AI/ML
18mArchitecting Intelligent Platforms
26mEngineering Excellence: Platforms and Architectures for Real-Time AI Analytics
31mUse-Case Driven ML Platforms
32mMaking Models Work: How an optimized ML lifecycle can drive better ROI