Privacy-Friendly Applications with Ollama, Vector Functions, and LangChainJS
summary
Claims a private stack that runs LLMs locally with Ollama, shows how embeddings are generated with Ollama and stored as vectors in SurrealDB, and demonstrates LangChainJS chaining that retrieves and composes those vectors for tasks like document summarisation and API orchestration. Demos include a vision model (moondream) producing image descriptions, code for embedding insertion and vector indexing, and a working LangChainJS vector-store integration that actually answers context-sensitive queries without sending data to external clouds.
AI-written summary · about
chapters (7)
- 0:00Introduction to AI privacy
- 2:32The case for local AI
- 5:58Getting started with Ollama
- 10:27Demo: Using Moondream
- 13:57Vector storage and search
- 21:15Orchestrating with LangChain
- 25:02Final demo and conclusion
related talks
31mBuild an Agentic RAG system using Langchain, Ollama and Milvus
01h 59mLLM applications with LangChain
29mUsing LangChain and LlamaIndex to customize LLMs...
27mBuilding Useful Apps With Small, Local LLMs
47mJava Meets AI: Integrating LLMs into Your Applications with LangChain4j
25mLangChain with Azure AI: Building LLM powered Applications
48mGoing Beyond Prompt Engineering
36mProduction Ready RAG