AI Blueprints are self service, pre-built, open-source ML pipelines that can be quickly integrated with any application
Today we’re excited to announce the public preview of Intel® Tiber™ AI Studio AI Blueprints. AI Blueprints are open-source pre-assembled machine learning pipelines that are built by Intel® Tiber™ AI Studio’s ML experts, but they can also be created for custom uses by any user. They’re for developers who want to build machine learning-powered apps and services for popular business uses like recommender systems, sentiment analysis, object detection, and many others. Developers can deploy a Blueprint as is or customize and run it easily. Get access to AI Blueprints within Intel® Tiber™ AI Studio Metacloud.
Why AI Blueprints?
Since we started Intel® Tiber™ AI Studio, our mission has been to give data scientists and AI developers the best possible experience for data science exploration and for managing sophisticated machine learning pipelines at scale. However, putting machine learning into use across the enterprise has been difficult due to the specialized expertise needed, as well as the orientation of machine learning tools to data science professionals.
Figure 1: AI Blueprints start with a simple mouse click
Making developers first-class citizens in machine learning
AI Blueprints changes this by putting developers front and center. Devs get all of Intel® Tiber™ AI Studio’s ML engineering expertise from end to end – data, preprocessing, training, deployment, and monitoring. With self-service ease, AI Blueprints lets developers choose from a collection of pre-built end-to-end pipelines that developers can use immediately. Create batch, streaming, or RESTful endpoints that serve your applications. Developers can even create streaming endpoints for serving real-time machine learning without the need to build pipelines from scratch.
Use off the shelf, or customize for your needs
Included in AI Blueprints is a comprehensive library of data connectors, pre-built ML pipelines and cnvrg ready-to-use blueprints. In addition to Intel® Tiber™ AI Studio’s own ML pipelines, you can build your own, and customize existing ones to meet an ever-expanding number of use cases. Use data from your own storage such as S3 bucket and Azure blob storage, or from your business Twitter account, Wikipedia or Salesforce. Run it through any of our AI Blueprints and get immediate results.
Figure 2: Retrain AI Blueprints for your own custom use cases
Ready to try it? Build your own Blueprints at our upcoming AI Blueprints virtual workshops
In these 2 hour workshops, you’ll use Blueprints and your own data to see how you can apply AI to solve real-world problems. These instructor-lead training sessions are hands-on, so you’ll walk away with an end to end ML pipeline and the knowledge to use or create new ones. Network with your AI peers and instructors to share ideas that you can implement right away. Register today here.
We’ll also be showing AI Blueprints in action at MLCon. Itay Ariel of Intel® Tiber™ AI Studio will show how to create a Slack chatbot in minutes using the new AI Blueprints. The session is on Tuesday, February 22 from 1:05 PM – 1:40 PM PST.
