TL;DR: It’s never been easier to build AI-powered Shopify apps with Gadget!

‍

New and innovative implementations of AI are popping up every day, from text-based chatbots and assistants, to photo generation and manipulation. Finding ways to incorporate AI-generated content into your apps is quickly becoming the norm, and now it’s easy to build AI Shopify apps with Gadget. 

Building the next generation

We’re always adding new features to help you skip boilerplate work that slows you down, like setting up infrastructure, connecting tools, or managing API versions. Last week, we launched some key features to make it easier to build AI-powered apps and tools with Gadget.

‍Node 18

Get access to any modern tooling that requires Node 18, like LangChain, web streaming, global fetch, and more.

Streaming responses

Use web streaming to return fragments of your responses to users, word by word, as soon as they are available.

Native vector data support

Most AI-powered chatbots, document services, resource searches, and other tools, make use of vectors. Now, this is supported natively in Gadget.

Focus on the interesting and challenging parts of your app build, like incorporating new technology or building the features and tools your customers ask for. Gadget still handles all the boilerplate and hosting for you, and brings AI tools to the table that allow you to go above and beyond your customers' wildest expectations.

New solutions vs old problems

With the new features and functionality, you can now create new tools and features, or provide a new approach to old problems. Now more than ever, buyers are looking for a shopping experience tailored to their wants, needs, and interests. Both sales and support can be supplemented through AI-based chatbots and assistants, product searches can be tailored to individual tastes. Before Gadget added the new vector data support, streaming responses, and latest version of Node.js, you would have had to connect to (and pay for!) one or more services in order to build apps that fully leverage modern AI features. Now, you can quickly get your AI-powered apps up and running with out-of-the-box support for what you need.

AI-based solutions don’t stop at personalization and support. Every merchant has run into the problems of pricing, logistics, reporting — the list goes on. There’s no shortage of ways that ecommerce can be supported and accelerated using AI, and although Shopify may be working on some solutions, they are unlikely to ever cover them all.

Building with potential

To make sure it really is easy to build AI apps with Gadget’s full-stack app platform, the Gadget team hosted an internal one-day hackathon, here are a few examples of what our team built..

An app that builds a data table component for you using Chakra UI

Building and formatting reports and tables for your data takes time, so our team took the opportunity to create a prototype app that builds and styles data tables using AI. Users simply describe what the table should look like, and the app will build it, supplying sample code that can be copy-pasted into Gadget’s hosted frontend.

An app that generates a simple frontend for interacting with your Gadget models

One team wanted to make building frontends in Gadget a breeze. They built an app that lets you explain what kind of frontend you want for your Gadget app, and the AI-powered chat will respond with React code that is already hooked up to your Gadget app’s API

An app that detects the primary colors used in uploaded images and presents image options to users when they select a color

For building color themes for collections or marketing materials, one member of the Gadget team created a tool that lets users pick a color, and any images that are deemed "similar" (using L2 distance comparison on the image’s vector embedding) are displayed. This was made easy to build by storing images in a Gadget app using the <inline-code>copyURL<inline-code> method of image storage.

A Discord support bot

For any tight-knit communities, it’s a game changer to have a chatbot that has all the information necessary to answer specific questions. Inspired by our own support team, one of the hackdays projects was the ability to ask questions about Gadget in a Discord community, and get a response pulled from the Gadget docs with a link to the docs page that the response was pulled from. Keep your eyes peeled for this one in the not-so-distant future.

Want to learn more about building AI apps in Gadget? Check out our building AI apps documentation. You can also join our developer Discord to ask Gadget employees and connect with the Gadget developer community!