Build an enterprise Agentic AI platform at Powercor
summary
Powercor deploys an extensible agentic AI platform on Amazon Bedrock using Claude Sonnet 3.5, a vector DB and hybrid RAG with reranking and critics to cut false positives in drone image reviews. It handles schematics, document retrieval and image analysis, routes work via a supervisor agent to specialist agents, and integrates email and field‑planning workflows. Concrete scale: 4 million customers across four networks, 90,000 km of conductor, 800,000 poles, 80,000 substations, ~200+ daily users and 2,000 calls/day; multi‑agent collaboration is production‑ready and speeds document access.
AI-written summary · about
chapters (6)
- 0:00Introduction to enterprise agentic AI
- 4:16What are AI agents
- 10:25Multi-agent collaboration benefits
- 16:59Powercor’s AI journey and demos
- 23:55Lessons learned and next steps
- 26:23Key takeaways and wrap-up
related talks
23mAgentic Backstage: How To Manage an AI Software Catalog
17mTransforming Robotics Service with Agentic AI on
43mArchitecting the Future: Modern Solution Design with Agentic AI
44mAgentic AI at Sage with AgentCore
31mAI Agent Orchestration for Enterprise Arch with AWS Bedrock, Flyte & LangGraph
30mLessons From Building and Deploying a Multi-Agent System at Scale: An...
18mAI Agents for a Green Future
16mTransform customer experience with an array of AI agents