From Containers To Cognitive Agents: Open Source Foundations for Enterprise-Grad...
summary
Claims an open source, container-first workflow that goes from local experiments to Kubernetes deployment using RamaLama serve for local LLM and RAG packaging, model pack (OCI) for bundling model+data+BOM, and model signing with OIDC (sigstore-style) for provenance. Demonstrates downloading a model, creating a RAG from PDFs, serving it locally, converting to a reproducible container image and pushing it to a registry, then verifying signatures. Notes 1.5M Hugging Face models with opaque provenance and ~50% of enterprises targeting agents by 2026.
AI-written summary · about
chapters (9)
- 0:00AI agents and enterprise adoption
- 3:14AI trust gap in enterprise
- 7:12Responsible GenAI Framework
- 13:56Model Openness Framework
- 17:53Local AI with Rama Lama
- 21:47Securing AI supply chains
- 26:09Evaluating agentic AI systems
- 32:32Demo: Rama Lama and model signing
- 43:08Closing and resources
related talks
47mRamaLama: Making working with AI Models Boring
39mBuilding Agentic AI Apps with Docker MCP & Cagent
23mBringing Container-Native Simplicity to AI/ML
32mOrchestrating AI Models in Kubernetes: Deploying Ollama as a Nati...
30mMind the Gap - A Developer's Roadmap To Building AI Agents
34mImplementing an open-source RAG with OpenSearch
01h 42mBuilding Responsible AI Agents in Decentralized Open-Source Environments
32mOperationalizing Openness: Standardizing AI Model Supply Chains With the Model Openness Framework