We understand the need for teams to be able to ingest ad-hoc data sources in addition to the regularized internal and external data inflows. In order to make this process efficient, Rubik Marvin provides uncomplicated channels for all platform users to retrieve new datasets and perform ad-hoc analyses. This is achieved by surfacing generic connectors that can be conveniently configured and re-used for data ingestion. This reduces the need to involve the product engineering teams in operational tasks and have them focus on more important core platform work.
We strongly believe that operational flexibility is what differentiates a good data platform from a great one. This starts with being able to immediately identify data quality issues and anomalies as close to the ingestion source as possible. The Rubik Marvin platform comes equipped with a suite of support tools to enable the ease of operations that are key to the success of any enterprise-grade software.
Different applications across an enterprise require diverse types of data stores to support specific business functions. A low latency user profile lookup service will need a high performance and low latency data store like Aerospike while an application that does financial reporting might need a data warehouse or an RDBMS.
Rubik Marvin uses a patent pending ‘pipeline as a source of truth’ approach to making this platform truly database agnostic. Users can write simple consumer jobs to retrieve data and its mutations at the most granular level and spin up any sort of database that is relevant to their specific applications.
When it comes to predictive accuracy, legacy data is as important as real-time data. The time travel feature allows data scientists and machine learning teams to go back to a specific point in time and retrieve the value of a particular attribute to allow for more effective model building and predictions.
The Rubik Marvin enables data access that is as easy as a Google search. Through a simple search format, users are able to ask any questions about the data and receive the right answers in return.