SOLVING THE CHALLENGE OF DELIVERING AI AT SCALE …
ELIPSA is excited to announce the release of our Rapid Deployment Template API! Now available in the Elipsa API library, our customizable templates enable customers to set up a single API to monitor any number of like devices. Just configure the template, connect a device, and we automate the rest.
In the land of Industrial IoT, we have met all sorts of characters from the successful plant manager who is interested in applying AI to increase productivity and reduce downtime, to the app developer who wants to predict water cavitation in her municipal water monitoring solution, to the OEM manufacturer looking to monitor thousands of machines at their customers’ sites to provide predictive maintenance.
As we have come to know these people as friends, we have seen one common characteristic, no matter how well they run their facility or organization, everyone is running at full steam with fewer resources than they had a year ago.
With this fact in mind, we started to think about how we can adapt our service to better serve our end-users at scale.
The barriers to AI adoption are not limited to data science; they also include the provisioning and deployment of AI into production. This proves especially challenging for customers that have hundreds or thousands of devices to bring online. Having already solved the data science challenge, we set out to solve the deployment challenge. How can we make it easier for our customers to realize the benefits of our AI engine without adding to their already busy workload?
After much discussion, some late nights, and helpful input from our customers and partners, we have delivered on that vision with the Rapid Deployment Templates. This is a giant leap forward in providing no-code, approachable AI, at scale.
To borrow a phrase from Ron at Ronco, we were aiming for a solution that would allow our customers to:
Elipsa’s Rapid Deployment Templates are the easiest and fastest way for an organization to scale automated AI across devices and use cases using one API.
Our templates allow users to configure a template that acts as a data pipeline to connect all of their like devices to one Elipsa API. Once the devices are connected to the Elipsa platform, we automate the creation, deployment, and management of the Elipsa AI models. This allows users to scale from one device to 1000’s quickly and without complication. Just point the new device, and we take care of the rest.
How does it work?
· Configure a template specific to your use case based solely on the data points that you want to monitor.
· Stream your data to the Elipsa engine using your template’s unique API endpoint. You provide the unique machine ID in your API call, and the Elipsa engine is able to automate the process end-to-end:
o Elipsa builds a historical dataset from your streaming data. Once you have streamed enough data, Elipsa automatically creates an outlier detection model.
o As you continue to stream new data, Elipsa will provide predictions based on the initial model, while adding to the historical data set.
o Elipsa then retrains the model on the larger historical dataset and makes all future predictions against the improved model.
Elipsa’s many-to-one approach through our Rapid Deployment Templates provides a true connect and play workflow that automates some of the most challenging aspects of scaling AI. Another leap forward in Elipsa’s mission of delivering fast, easy, scalable AI for any workflow.
Where do you go from here?