Abstract
An internal tech talk exploring the landscape of locally-run AI agents — running large language models on your own hardware using tools like Ollama, building agentic workflows without sending data to the cloud, and the practical trade-offs between local and cloud-hosted AI for enterprise development.