The Versal AI SDK is a TypeScript-first toolkit for building AI features. It streamlines text generation, embedding and structured output.
We’ve just posted a course on the frequedcamp.org YouTube channel that will teach you how to use Versal AISDK to create and deploy a customer support agent that makes autonomous decisions to either answer questions based on your support documentation or search the web in real time.
In this course, you will deploy customer support agents who:
Embeds support documents in the Sopas vector store.
Uses retrieval and web search as tools, selected on the fly based on the user’s query.
Classifies intents with structured output (via GenerateObject + Zod).
Answers to questions with basic, reliable responses – Connect to your documents when looking for relevant or web searches at the desired time.
These topics are covered in the course.
Define chords and embeddings and decide when to use each.
Configure Soupbase as a vector store: create tables, embed documents, and handle chunking/text splitting for large files.
Implement retrieval with Soapbase RPC so that your agent fetches the correct context for any query.
Use the Versal AI SDK Basics: Embed and generate text for fast, reliable model calls.
Generate structured output with GenerateObject and Zod for optimization and root intents.
AISDK-define schemes, call tools with string implementations, and keep everything type-safe.
Treat retrieval and web search as tools, and compose them into a single agent decision flow.
When your documentation isn’t enough, use the OpenEye web search tool to pull up fresh, real-time information.
Combine all of this into a support agent that chooses the best strategy (recovery, search, or direct response) and explains its responses.
View the full course on the freecodecamp.org YouTube channel (2 hour clock)