Deno's batteries-included JavaScript/TypeScript toolchain + Jupyter = simple way to work with LLMs

View source here: https://github.com/kitsonk/deno-llm

Read the accompanying blog posts:
https://deno.com/blog/the-dino-llama-and-whale
https://deno.com/blog/build-custom-rag-ai-agent

00:00 Intro
00:36 Fundamental concepts
01:00 Setting up
01:50 Building RAG agent
02:40 Pull new models
03:02 Process, parse, convert blog content to vector memory
03:42 Graph of RAG agent steps
04:12 Creating each step
05:25 Putting steps together
05:57 Testing the RAG agent


Resources
- Ollama: https://ollama.com/download
- Deepseek: https://deepseek.com
- Jupyter notebooks: https://jupyter.org


Deno is an all-in-one batteries included JavaScript and TypeScript toolchain.

Website: https://deno.com
GitHub: https://github.com/denoland
Discord:  https://discord.gg/deno
Bluesky: https://bsky.app/profile/deno.land
Mastodon: https://fosstodon.org/@deno_land
Twitter: https:/twitter.com/deno_land

Deno's official blog provides updates, tutorials, and use cases for Deno, a secure JavaScript/TypeScript runtime. Developers can explore Deno's features, security model, and ecosystem integrations, gaining insights into modern JavaScript development practices. With contributions from Deno core contributors and community members, the blog offers resources for developers adopting Deno for their projects.

Deno

Discover the simplest approach to integrating large language models (LLMs) into applications using Deno and Jupyter Notebooks. Learn how to install Ollama for local LLMs, use the DeepSeek R1 model, and leverage the LangChain library for consistent interaction with LLMs. Additionally, explore the creation of a Retrieval Augmented Generation (RAG) agent for AI chatbots with domain-specific knowledge, and see a step-by-step process of building and deploying such an AI agent effectively.

The easiest way to work with LLMs