August 2019 Release: Iris AI enhancements

In this video, learn how SnapLogic platform lower the barriers to entry for citizen integrators and decreases time to value for integration specialists using Iris AI enhancements.

Read full transcript

This demo is focused on SnapLogic Iris AI Enhancements. First, I will focus on Enhanced Search capabilities and then talk about another new innovation called “Pipeline Recommendations.” 

With the improvements we have made to Snap Search, when you search for a keyword that does not match any Snap or Snap Pack name, Iris now provides suggestions based on the content on the SnapLogic documentation site.

If I say “csv”, it will do a name match and return the relevant snaps that I can then use to build my pipelines. 

If I want to build pipelines using sftp, let’s see what happens when I use that as a search term. It is able to retrieve a list of snaps where the word “sftp” was referenced on the documentation page. 

Let me show you the documentation page for the File Reader Snap where the word sftp shows up and that’s why it got indexed and showed up in the search results. 

Now let’s do another search using the search term “marketo” and this will now return the Rest Snap. That is because we have referenced “marketo” on the Rest Snap documentation page as Marketo REST API can be used with the Rest Snap Pack to build integrations with Marketo. 

With these improved search capabilities, we now have ad-hoc or citizen integrators enabled to use common terms to find relevant Snaps to build their pipelines thus improving their productivity. 

With the improvements we have made to Pipeline Search, when you search for a Pipeline, Iris now highlights the projects spaces where it finds a match and shows you a count of the matches against each of them. 

Let me search for pipelines that process CSV files. Now let’s navigate to the SnapLogic Data Science project and look at the CSV To JSON Conversion pipeline by opening it. This looks like a good example I can use for CSV handling. 

Now let me talk about another new innovation called Pipeline Recommendations. Two years ago, SnapLogic was first to bring AI to enterprise integration with the introduction of Iris AI. At launch in May 2017, Iris AI applied machine learning to predict the next likely step, or Snap, in building a pipeline. Now, with the August 2019 release, the Iris AI recommendation engine takes this further by suggesting fully-completed pipelines, from original source to end target, and all Snaps in between. 

These suggestions are based on existing pipelines residing within the user’s organization and pipeline patterns from the SnapLogic Patterns Catalog. When a user identifies two endpoints they want to connect – such as Salesforce and Workday – Iris AI will build the entire pipeline in real-time. It does this by leveraging advanced algorithms that have learned from the data flows and integration paths from previous successfully executed pipelines. From there, the user can adjust the pipeline Snaps if needed. This AI-assisted head start shortens the learning curve for new users and boosts productivity for expert integrators.

Now say I wanted to do some work with Postgres where I read from a file, do some transformations and write to Postgres. So, let me drag the File Reader Snap and the Postgres Insert Snap onto the canvas. 

Let’s click on Suggested Pipelines to look at various types of recommended pipelines. I can preview each one of them and if I like one, I can open it in my canvas. 

So, let’s pick the first one and replace the current pipeline. I can now go ahead and configure the pipeline but what Iris has done here: it has generated a complete pipeline. 

Now let me show you a pipeline that has the word ‘Salesforce’ in it. The suggested pipelines list other pipelines dealing with in the organization as well as relevant patterns from the SnapLogic Patterns Catalog. 

Looks like we have an interesting pattern that integrates contacts from Salesforce to ServiceNow, so let me pick that one. 

The key benefit here is improved productivity for less technical users – Citizen and Adhoc Integrators as they can now leverage entire pipeline recommendations instead of just snap recommendations. 

Thank you for watching this video. If you would like to know more about SnapLogic Iris AI, please visit


The AI Mindset: Getting Started with Self-Service Machine Learning

Watch Now

Analyst Report

Gartner names SnapLogic a Leader in 2019 iPaaS Magic Quadrant…

Read Now

Case Study

Making the faculty and academia even smarter at Boston University

Read Now
Contact Us Free Trial