Build your own Autonomous Agent powered by LLMs using Kubernetes
summary
Builds a production-ready autonomous agent pipeline that scrapes docs and Slack, vectorizes content into a vector store, then spins up an agent that answers documentation queries and can be deployed to Slack or a website. Uses LangChain for agents and tools, ZenML to define a four-step pipeline (URL scraper → web loader → index generator → agent deploy), and Kubernetes to run scheduled, resource-limited jobs and host vector DBs. Demonstrations use Dolly/Stability models and open-source components; pipelines run successfully on Kubernetes stacks.
AI-written summary · about
chapters (9)
- 0:00Introduction and context
- 1:06Understanding LLM agents
- 2:36Concepts of agent design
- 5:59Agent architecture blocks
- 9:58Frameworks for LLM agents
- 13:37Pipeline requirements
- 18:51Building with ZenML
- 21:28Kubernetes and infrastructure
- 27:24Code and pipeline demo
related talks
25mLLM Powered Agents with Kubernetes
42mLLMs on Autopilot: Running AI Agents on Kubernetes With Open Source Tools
40mAgents in Action: Advancing Open Source AI Missions
23mAgents in Kubernetes 101: A Crash Course on Cloud Native AI Agents
31mBuilding and Orchestrating Real Production-ready Agentic AI Systems
38mBuilding AI Agents and Assistants for DevOps Workflows
22mHOW TO BUILD YOUR OWN AI AGENT: THE COMPLETE ARCHITECTURE FOR ENGINEERING TEAMS
19mBuilding Agents at Cloud Scale