We've seen how Databricks Serverless SQL makes a huge difference for serving data out to reporting tools in a fast, cost-effective way, but how do we do the same with online inference? How can we easily standup a webservice that can scale to meet huge demand, without paying a huge cost. Today we're looking at Databricks Serverless Realtime Inference, currently in public preview.
In this video Simon is one again joined by Gavi, our resident Data Science expert, to take a look at the new feature, try standing up a serving endpoint and check out some of the available features!
For more info on serverless realtime inference in Databricks, check out: https://learn.microsoft.com/en-gb/azure/databricks/mlflow/serverless-real-time-inference
There's also a deep-dive Databricks webinar coming up around Production ML at Scale - register here: https://www.databricks.com/resources/webinar/enabling-production-ml-scale-lakehouse
And for our online, self-paced training, come visit us at: https://advancinganalytics.teachable.com/