Jay Kreps: I definitely think that there’s an aspect of us just figuring out how to do things more efficiently and willingness to make adjustments faster in that respect. There’s obviously areas where there’s trade-offs. And so — nothing in life is free. But — like, yes, we felt that we were able to make this change without significantly changing. Now I would say, look, there is something impacting growth, which is — we are in a macroeconomic environment that’s very different from a year ago, and that’s a headwind. And so I think when we were considering what we were going to do on the expense side, we were taking into account that we were going to be facing this headwind and likely growing slower than we would be if that was not the environment that we were operating in.
Howard Ma: Okay. Thanks. And I just have a quick follow-up for Steffan. On the platform side, Steffan, I forget if you mentioned this in your prepared remarks, but if — or I might have missed it. But was there any notable change in contract duration on the platform side that resulted in lesser license revenue recognition than in prior quarters? And also, is there any migration from the platform to Cloud that’s worth calling out? Thank you.
Steffan Tomlinson: Thanks, Howard. There was no material change in contract duration. But what you’re seeing drive the change in license revenue is really the profile of new ACV that’s coming in the door. And the new ACV is Confluent Cloud. It was very, very healthy this quarter. And we saw just less new platform deals come in because as the industry is all heading towards cloud. With that said, Confluent platform is still an important part of our portfolio. And we’re going to continue to see contribution from platform, but it’s really about — like, Cloud is the story here. And even going back to a prior comment that was made, even in a tougher macro environment, we just came off of a quarter where we posted record Cloud sequential growth. And we’re calling for very meaningful Cloud expansion over 2023. And that goes back to the testament of the value that we’re delivering in our Confluent Cloud model.
Howard Ma: Great. Thanks guys.
Steffan Tomlinson: Thank you.
Shane Xie: We’ll take our next question from Derrick Wood with Cowen, followed by Wells Fargo.
Derrick Wood: Great. Thanks for taking my questions. I guess, first, Jay, I wanted to touch on the Immerok acquisition. What does Flink excel at that improves upon the capabilities of Kafka streams or ksqlDB? And how should we think about maybe the R&D shift as you bring Apache Flink in? Are there some technologies that you’ll look to deemphasize going forward? Or what’s the balance across the stream processing technologies that you have?
Jay Kreps: Yes. Yes, it’s a great question. So — yes, Kafka Streams is effectively — it’s an application development library that helps you do stream processing with Kafka. So it’s very easy to use in embedded applications. It tends to serve more kind of micro service use cases. What Flink brings to the table is, I think, really the most complete, well-thought-out framework for stream processing, it generalizes batch processing with real-time streaming. So you can kind of run thing, something at a point in time and then have it keep running up into the future. It supports a variety of programming languages, so Python, Java, SQL. It has probably the best scalability and performance. It has, I think, the most active community.
