Serverless Wasm Functions Workshop

This repository contains resources for a half-day workshop focussing on Serverless Wasm Functions. In addition to a high-level introduction, the workshop consists of three Hands-On-Labs (HOL) exercises. We highly encourage workshop participants to code along and get a sense of how streamlined a developer experience (DX) for building truly serverless edge-native applications feels like in 2026.

Prerequisites

Workshop participants should bring:

• A GitHub account (The GitHub account is used to authenticate against Serverless Wasm Functions)
• An editor or IDE
• Permissions to install software
• Permission to allocate a local port (by default 3000) for running and testing applications

If you're not participating an instructor-led version of this workshop, access to Akamai Cloud is required for provisioning a Linode including a dedicated GPU to run large language models (see Hands-On-Labs 3)

Suggested Running Order

1. Start with the Introduction to explain core values and benefits from using WebAssembly to build and run serverless workloads. The presentation provides an high-level introduction to technical underpinnings, introduces Spin and Serverless Wasm Functions.

2. Workshop participants setup their development machines by following the language- and os-specific instructions.

3. Hands-On-Labs 1: Build & Deploy in 4 Steps is an interactive exercise to familiarize yourself with the developer tooling (spin CLI) for building and running WebAssembly applications. As part of this Hands-On-Labs, you will also create your account with Serverless Wasm Functions and deploy the first application.

4. Hands-On-Labs 2: Edge-Native Use Cases is another interactive exercise which teaches fundamental building blocks of Spin and Serverless Wasm Functions. You will make use of language-specific Spin SDKs to address common day-to-day concerns when building edge-native applications.

5. Hands-On-Labs 3: Edge Accelerated Generative AI is the final exercise illustrating how you can build smarter applications for the edge by adding generative AI to the mix.

Hands-On-Labs folder structure and checkpoints

Although Spin supports many different languages, samples are currently provided in Rust and TypeScript only.

With Hands-On-Labs 1, attendees get a chance to start from scratch and learn the "Spin Workflow". Other Hands-On-Labs folders always follow this structure:

hol-folder/
 | - README.md
 | - start-rust/
 | - start-ts/
 | - solution-rust/
 | - solution-ts/

Workshop participants could use the corresponding "checkpoint" (solution-rust or solution-ts) if they have problems with completing a particular Hands-On-Labs. Consider these solution folders as "checkpoints" allowing you (workshop participant) to have the flexibility of extending Hands-On-Labs individually or diving straight into more advanced topics.

Supportive Cloud Infrastructure

Hands-On-Labs 3 (GenAI at the Edge) requires a Linode (with dedicated GPU) being provisioned and at least one large language model (LLM) being exposed through Ollama. For instructor-led workshops, workshop participants could expect this infrastructure being pre-provisioned and ready-to-use. (HTTP endpoints will be provided by the workshop instructor)

The Cloud infrastructure setup instructions could be found in the infra folder.

Cleaning up your Serverless Wasm Functions Account

As part of this workshop, you will deploy several applications to your individual Serverless Wasm Functions account. To retrieve a list of applications deployed to your account, can always use the spin aka apps list command.

To delete applications deployed to your account, run the following commands:

spin aka apps delete --app-name hello-serverless-functions --no-confirm
spin aka apps delete --app-name edge-native-use-cases --no-confirm
spin aka apps delete --app-name edge-accelerated-genai --no-confirm