Build a RAG based AI coding assistant for your open source project using Wasm [Workshop]

Michael Yuan - Second State

Generative AI and Large Language Models (LLMs) are revolutionizing the developer experience, and coding assistance is one of the most promising applications.

In this workshop, tailored for Wasm developers and contributors, we will explore how AI-powered coding assistants can:

• Seamlessly integrate with leading commercial IDE tools.
• Help experienced developers contribute more effectively to open source projects.
• Accelerate onboarding for new developers and users for your projects.
• Assist project maintainers in reviewing and managing community contributions.

Then, you will learn how to create a knowledge agent for your own GitHub repo or documentation site. The agent consists of an open-source LLM fine-tuned for coding tasks, specialized prompts for your project (ie the LLM.txt file) and vector knowledge based based on your code repo and docs (ie the RAG or Retrieval Argumented Generation approach). All the application is built with Rust and Wasm.

At the end of the workshop, you will be able to set up a coding assistant for your own GitHub repos and make it available to your community.