Serverless Architecture for Machine Learning
summary
Presents the team’s evaluation of serverless architectures for a machine learning solution, comparing choices and trade-offs rather than only a final design. Covers the user interface layer, workflow/orchestration layer, processing layer, cloud pricing strategies, and team/process/cultural aspects, with emphasis on practical decisions informed by experience in databases, data engineering and data science.
AI-written summary · about
chapters (9)
- 0:00Introduction
- 3:12Client context and challenges
- 7:25Serverless architecture proposal
- 12:23Tools and components
- 18:01Model training process
- 23:45Production integration and feedback
- 29:43Model evaluation and metrics
- 33:25Real feedback importance
- 36:48Final considerations and next steps
related talks
26mDeploying Machine Learning Models in Serverless Architectures
43mServerless architectures — approaches to building
52mWhy Serverless?
46mServerless approach to software architecture and development
42mNo Server Is Better Than No Server: Building Blocks for Serverless Architectures
41mIs Serverless the End of Kubernetes?
01h 00mProcessing a Data Lake on Node.js in a Serverless Architecture
18mServerless AI's Rise: Revolutionizing ML Deployment