DataRobot adds automated AI reports, Snowflake integration

DataRobot, a platform that aims to bring AI technologies to enterprises, today announced its second major platform release, DataRobot version 7.1.  With the introduction of MLOps management agents, time series model enhancements, and automated AI reports, the goal is to enable organizations to drive business outcomes with AI and accelerate customers’ time to value, the company says.

Almost a third of organizations are using some form of AI, with 43% reporting that their rollout accelerated as a result of the pandemic., according to a recent IBM report. Adoption is being driven by both pressures and opportunities, from the pandemic to technological advances that make AI more accessible.

DataRobot version 7.1

Arriving in DataRobot 7.1 are MLOps management agents, which provide lifecycle management for remote AI and machine learning models. Management agents understand the state of models regardless of how they’re created or where they’re running and can automate tasks, including the retrieval model artifacts (i.e., outputs created by the training process) and deployment or replacement of models directly in their environment.

DataRobot 1.7 also offers feature discovery push-down integration for Snowflake and time series Eureqa model improvements. Leveraging this, Snowflake customers can tap into automatic discovery and computation of features (the individual independent variables that act as the input in a model system) in the Snowflake data cloud. They can also run forecasting models in Eureqa, the proprietary modeling engine originally created by Cornell’s Artificial Intelligence Lab.

Eureqa models are based on the idea that a genetic algorithm can fit different analytic expressions to trained data and return a mathematical formula as a machine learning model. Genetic algorithms simulate the process of natural selection, generating high-quality solutions for optimization and search problems. As DataRobot explains, with smart feature selection, Eureqa models can reduce complexity and work well with both large and small datasets.

Lastly, DataRobot 1.7 provides a no-code app builder that allows customers to turn deployed models into AI apps without coding. The app builder can score new data, perform what-if scenarios, and run simulations to identify input values that might optimize an outcome. Complementary capabilities include automated data prep for time series, “nowcasting” for time-aware models, reports that automatically summarize findings of modeling projects, and prediction jobs and a scheduling interface to manage and maintain prediction schedules in one place.

“We are in constant communication with our customers regarding the challenges they face when deploying AI and as a result will tailor our updates based on their unique needs,” DataRobot SVP Nenshad Bardoliwalla said in a press release. “We’re thoroughly committed to creating a platform that empowers every individual — from the most advanced data scientists to the everyday, non-technical business user — to take advantage of AI. By easing the model lifecycle process and cutting down time to value, this latest round of enhancements gives enterprises the tools they need to better build, manage, and see value from their AI projects.”