PennyMac Secures Sensitive Data and Saves Costs with Snowflake

This transcript was automatically generated.

I'm Cyrus Tibbs. I am the chief information security officer at PennyMac. I think PennyMac's vision is to be a top trusted lenders and, you know, producing loans, servicing loans, and creating, investment vehicles out of them. So our vision is to help people get into their homes.

As a mortgage company, we collect some very sensitive data on our customers that we need in order to run our business. As the, CISO, my my job is to protect that data, but also enable the business to use it in responsible ways. Snowflake enables me to essentially take a best of breed approach where I can buy the best signal. I can buy the best endpoint signal, the best email protection, the best network protection.

And then I can take the data between all those different systems and bring it into Snowflake. And where I need to have that alignment between my data systems, I can use Snowflake to build that. I mean, Snowflake gives you that optimization at the detective side. Those are dollars you can reinvest in the preventative side, and you'll get a lot more value out of that when you take that approach.

Hunters is a fantastic partner for us. And given that we are, you know, a hundred percent in cloud, Hunters really is sort of that team that handles our detection engineering, like all the major things we need to detect and respond to that are kind of undifferentiated. And then in addition, Hunters takes our data and drops the raw feed into a shared Snowflake database, where then my team can build our own data models.

We plug Streamlit into it to build our applications. We can focus on building where it's, value producing for the company. I would say in the past, the legacy SIEM would make up sometimes, you know, I'd say thirty to forty percent of the total cost of a security program between the the ingestion based costs and the people to maintain it and the data engineers and sort of what it would cost for a security team to maintain their own data silo. So now, you know, my, you know, security data model and the all the technology people around it represents more around, like, the five or ten percent of my overall budget. I think this is the first time in my career where I'm not worried about the cost of adding new data to my security data lake. I'm essentially trying to take as much data in because I wanna make sure that whatever new capabilities are out there, that I have the right quality data to take advantage of them.