Source training data and actuals
Connect
SPIRITT
DataRobotDataRobot
Models in production, owned
SPIRITT runs the full DataRobot lifecycle: data in, Autopilot, deployment, batch scoring, drift and accuracy monitoring, retraining and documentation, with alerts that wake it when something changes.
What you can do with DataRobot.
From data to deployed model
Create datasets from your sources, run Autopilot, review cross-validation and feature impact, and deploy the winner.
Monitoring that acts
Track drift, accuracy and service health, and route DataRobot alerts to a webhook that wakes SPIRITT to respond.
Predictions where work happens
Schedule batch prediction jobs, upload actuals and keep scores flowing into the warehouse and apps that use them.
Works with your other apps.
Get started in three steps.
01
Open your workspace
Start a SPIRITT workspace for your ML operations, with your data sources, models and the teams that consume predictions.
02
Connect DataRobot
Connect DataRobot with your API key so SPIRITT can work with projects, deployments, predictions and monitoring.
03
Hand over the lifecycle
Describe the prediction problem or the deployments to own. SPIRITT builds, runs and watches them.
Hand over a mission.
Choose a mission. Make it yours.
Put DataRobot to work.
Every model, built and watched.
Questions
DataRobot, with SPIRITT.
01What can SPIRITT do with DataRobot?+
Create datasets and projects, run Autopilot, inspect model insights, deploy and replace models, schedule batch predictions, upload actuals, monitor drift, accuracy and service health, set retraining and notification policies, and evaluate GenAI blueprints.
02Can DataRobot alerts start work in SPIRITT?+
Yes. SPIRITT creates a webhook notification channel and policies for events such as accuracy or drift changes, so alerts reach SPIRITT and it starts diagnosing right away.
03Can it keep predictions flowing into our warehouse?+
Yes. SPIRITT sets up batch prediction job definitions that read from and write to your data stores on a schedule.
04Does it handle model governance?+
Yes. SPIRITT generates compliance documentation, manages sharing roles and records model replacements and retraining outcomes.
05Does it support generative AI work?+
Yes. SPIRITT creates playgrounds, LLM blueprints and test suites, compares blueprints side by side and configures moderation guards.
06How do I get started?+
Open a SPIRITT workspace, connect DataRobot and your data source, then name a prediction problem or a deployment to own. SPIRITT starts there.