This post is one in a series of quick-hit interviews with companies using DataStax Enterprise (DSE) for key parts of their business. In this interview, we talked with Ole Pedersen, who is in charge of Research and Development at Raptor.
DataStax: Hello Ole, thanks a lot for your time today. Can you tell us a bit about Raptor and your role at the company?
Raptor: Raptor Services A/S is a true Omni-channel market leader within behavioral recommendations using advanced algorithms, data mining and machine learning. Our data platform collects, learns and recognizes behavioral patterns for each user. This empowers marketers to provide relevant content based on consumer behavior across all platforms and channels.
I am a Senior Data Scientist at Raptor Services A/S, and I am responsible for “Research And Development.” I am also the database administrator for our Cassandra cluster and responsible for development of our data mining and machine learning algorithms on the DSE platform.
DataStax: What makes your personalization solution successful and what differentiates you to similar applications?
Raptor: Our success is due to the behavioral understanding of processed data which makes it possible to create individual customer profiles down to a 1:1 scale. Using behavioral recommendations, Raptor Smart Advisor creates memorable user experiences with personalized recommendations of relevant products in real time. Customers will increase basket size, and they will return for more because offers provided to them become more relevant over time. This will increase customer experience and satisfaction rates.
All of this is done in real time, and in a highly customizable solution, which makes Raptor the perfect recommendation platform for a wide range of customers. The speed, accuracy and flexibility differentiates us from competitors.
DataStax: Did you use a different technology before you started using DataStax Enterprise (DSE)?
Raptor: Yes, we are migrating from a setup where we were running batch jobs on MS SQL Server.
DataStax: Why did you decide to use DataStax Enterprise? What kind of data is stored in DSE?
Raptor: As our company has grown, we have gotten more data than we could handle in the SQL Server setup. To keep up with the demand, we looked for better alternatives. We also wanted to offer a real time experience to our users. This led us to begin storing user behavior and data mining models in DataStax Enterprise.
DataStax: How would you sum up the benefits you’ve achieved with DataStax Enterprise?
Raptor: In DSE (or Apache Cassandra), we have a platform where we can easily scale linearly by adding nodes. The database engine is also widely used, well documented and supported. It provides a feature set matching our needs well, including, but not limited to, TTL expiration of old data, counters, wide rows.
DataStax: What caused you to use DSE over open source Apache Cassandra™?
Raptor: Actually, we started out with Apache Cassandra but switched to DSE to get a tested and stable platform, to have the ability to monitor the cluster through OpsCenter and also for the benefit of DataStax’s free support offer to startups.
DataStax: What features from the DataStax Enterprise stack are you using at the moment? What business / customer experience outcomes have you achieved by using DataStax Enterprise?
Raptor: We are using DSE for reading and writing user behavior, and we use Spark to batch update the data mining models. The platform is central in our stack.
We also use OpsCenter to monitor the cluster. Monitoring the cluster over time enables us to spot trends in, for example, read request latency or OS load. We can then act on them, which lets us provide a stable platform for our customers.
DataStax: Tell us about the future of your project(s), do you intend to leverage other parts of DSE to make it a reality?
Raptor: We plan to expand our set of real time and batch algorithms using the Cassandra and Spark parts of DSE. We are not looking into other parts of DSE right now, but that could change as we expand.
DataStax: What advice would you give to other startups that are thinking about using Apache Cassandra™ and DSE for the first time in their solutions?
Raptor: If you are new to Cassandra, be sure to test for scale before hitting production. The way you use Cassandra will determine if your setup can scale. Read up on do’s and don’ts. After hitting production, be sure to monitor your cluster’s performance over time using OpsCenter so you can see how data size and load increases.
SHARE THIS PAGE