The Sensai Platform

The Sensai Platform is an Artificial Intelligence platform that was created as the product offering of Sensai Corporation.  The Sensai Platform was acquired by Sovereign Intelligence in September of 2016.

My Role


At Sensai Corp, I acted as VP Product and Cofounder.  It was a role that also oversaw not only the Design Team but also a team of five engineers.  Mainly, I did my best to make the voice of the customer heard in all things while balancing design, feature, and user experience issues.  Other duties

The Platform​: CDE

Sovereign Intelligence’s Sensai Concept Discovery Engine (CDE) is a platform designed to work with the ElasticSearch/Solr stack.  You can use the CDE to convert plain-text information in your logs or documents into structured, quantifiable data that enriches the already powerful data foundation ElasticSearch provides.  The CDE uses a myriad of Artificial Intelligence techniques to create a smart pattern analysis that allows you to get at the un-analyzable data locked in your documents or logs.


A traditional ElasticSearch use case often requires analysis of both structured and unstructured data.  For example, a log-analysis system has many structured fields: Dates, times, percentages, load values, error types, and more.  However, the text message portion of log entries is often left unanalyzed, or simply matched against a small assortment of interest keywords. Being able to convert the prose of the message into clean, quantifiable data greatly expands the value of your logs. Having the error type, code location, and other clues that programmers leave in the text portion of the message available alongside the discrete quantified data makes the analysis much more valuable.


Other use cases such as news analytics, competitor awareness, market analysis, and business intelligence in general also share this mix of quantified and unstructured text data. Applying a pattern-based extraction system like CDE to the textual data allows you to extract names, events, sentiment, and relationships that would otherwise go missing. That information can be moved into the same analytic domain as the quantifiable data where it can be sliced and diced with all the powerful analytic tools already available in the X-Pack and beyond.

© 2017 by MICHAEL P. GUSEK

San Francisco, CA | | 415.259.1500