Domino and DataRobot Deliver More Value Together

March 1, 2018
· 4 min read

Today’s Artificial Intelligence (AI) solutions projects require a lot of collaboration and teamwork in order to grow and succeed. However, limited resources — such as not having enough available and skilled data scientists and tools that don’t offer high levels of automation — are hampering even the best teams out there. This is the moment when partnerships and collaborations step in to resolve these limitations as companies join forces to provide the resources needed to ensure success. It’s like that saying goes, “Two heads are better than one.” Two platforms working together are better than one standing alone.

This collaboration exemplifies how two companies can achieve balance and can work in sync on shared projects and opportunities. 

DataRobot has various partners from around the world, and values these connections. One example of an effective and successful partnership is with Domino Data Lab. Together, DataRobot and Domino offer a unique combination of tools that empower AI teams, data scientists, and other domain experts to be more productive and provide the resources needed for project success. This collaboration exemplifies how two companies can achieve balance and can work in sync on shared projects and opportunities. They feed off of each other’s strengths and successfully join forces based on their commonalities.

Predictive Modeling Made Easier

One major limitation for AI applications is creating models by hand, which is a time-consuming and challenging process that hinders the progress and productivity of many companies. Domino’s patent-pending reproducibility and collaboration engine offers you a collaborative coding environment where your data science team can work together to prepare the data, develop or work with important models, and deploy your models.

Next up is DataRobot. We deliver automated machine learning, which helps your team quickly develop dozens of models to consider while providing transparency for understanding how these models work. The advanced transparency and visualizations built into DataRobot enable skilled users to see exactly how a certain model performs the way it does, and how to get a better understanding of the underlying data. DataRobot ranks all of the models by accuracy on its leaderboard, allowing you to quickly evaluate the models. DataRobot’s advanced automation ensures that you can find the best model to fit the task at hand.

Several major enterprises are already using Domino and DataRobot together. 

Once you’ve selected a model, there are multiple deployment options from which to choose. DataRobot offers three options (see How to Deploy AI Solutions to Production), and Domino allows API endpoint publishing (see Domino site for more details). Domino also provides additional project collaboration and sharing, which is a good way to wrap up the project with input from everyone involved. Bringing the selected model from DataRobot into Domino provides the opportunity for more collaboration and further refinement.

The end-to-end workflow when working with Domino and DataRobot:

  1. Prepare data, do manual feature engineering, and build models in Domino
  2. Send that data to DataRobot to automatically build, validate, and tune more models for the user to consider
  3. Model validation and model insights can be done in both DataRobot and Domino
  4. Models can be compared and the best, most accurate model is then deployed with DataRobot, or through Domino for additional project collaboration and sharing
  5. Model predictions can be turned into visualizations and business decisions using Shiny and Flask apps on Domino, and the full pipeline can be deployed in Domino, which is available via API


Several major enterprises are already using Domino and DataRobot together. Domino allows data scientists at these organizations to write scripts for pulling the data, feature engineer, and then either save the file or use the DataRobot API to programmatically build models using DataRobot. Data scientists can then explore a leaderboard of possible models within DataRobot, discovering significant patterns revealed in the data. They can also tweak how the predictions are turned into business decisions and create insightful visualizations.

From there, data scientists can iteratively hand-tweak the best models in Domino’s coder-friendly notebook environment, which provides unlimited library support, versioning, and full reproducibility — a true end-to-end platform for data science.

Together, Domino Data Lab and DataRobot enable data scientists to accelerate the development and delivery of models. Through this collaboration, users of both solutions are empowered with key capabilities of rapid model development, automated machine learning, seamless collaboration, and one-click model deployment. This greatly increases productivity and removes bottlenecks in the modeling process.

Check out this simple introduction to how a Domino user can take advantage of the DataRobot automated machine learning platform.

 New Call-to-action

About the Author:

Dan Ganancial leads Partner Marketing at DataRobot, and he is responsible for driving joint marketing initiatives with alliance and channel partners. Dan is a marketing professional with more than 10 years of experience in partner, product, and strategic marketing. He has held several roles in his career related to consultative sales, business development, and marketing where he has produced a strong record in driving both customer and revenue growth. Follow him on Twitter – @datarobotdan

About the author
Dan Ganancial
Meet Dan Ganancial
  • Listen to the blog
  • Share this post
    Subscribe to DataRobot Blog
    Newsletter Subscription
    Subscribe to our Blog