Building an AI-Ready Data Foundation
summary
Platform-level walkthrough of how an operational AI foundation is built: Tines runs 1 billion weekly actions on a workflow platform that integrates AWS Bedrock plus PrivateLink to guarantee tenant-scoped, region-bound processing with no training storage logging by default; AWS supplies GPU/Trainium infrastructure (P6, Trainium3), Bedrock model choice, OpenSearch GPU-accelerated vector search and S3 vectors, Agent Core to deploy agents in under a minute, near-instant DB creation and a database savings plan that yields up to 35% savings.
AI-written summary · about
chapters (7)
- 0:00Introduction to AI-ready data
- 2:35Tines AI workflow platform
- 11:02Building and deploying agents
- 15:54Agentic AI data challenges
- 26:30Effortless data management
- 39:50Agent deployment and frameworks
- 42:59Summary and future outlook
related talks
35mHigh-performance data foundation for AI success
24mBuilding GenAI applications with Amazon Bedrock
20mFoundation Models for Big Data and Prompt Tuning
20mData is Your Differentiator: Building Secure and Tailored AI Systems
47mBuild cloud data management across analytics, ML & gen AI applications
16mEnterprise AI Innovation: Gen AI Experimentation
16mEnterprise AI Innovation: GenAI Experimentation
38mGenerative AI - Architectures and applications in depth