Driving subscription growth via ranking/classification using deep neural networks on Tensorflow

Location intelligence using machine learning (for ranking & classification)

- Core team member since inception, led the technology R&D across the company's product lines* developed a rule engine to filter `interesting' transactions in a bank, which led to an anti money laundering product* developed a business intelligence product, where a financial institutions' relational data is imported into a SSAS cube, and analyzed in a pivot table built in Flex* developed a C# API to emulate a versioned file-system, with pluggable storage support; in particular, a file-system and a relational database back-ends are ready* oversaw the development of a loan origination system, where potential customers were identified and clustered based on various financial, demographic and personal parameters* undertook consulting assignments on behalf of neorithm

- Core team member, enabled rapid application development with an in-house framework* developed an enterprise application development framework, consisting of workflows, an ORM framework, message queues; and code generators for an ASP.Net based UI* assisted in the development of an ATM switch, which could be theoretically support any ATM protocol; in particular, the Diebold and NCR protocols were developed* played a major role in improving the usability of web applications, using JavaScript to bring desktop-like features to web applications

- Implemented Amdocs' telecom billing functionality using Unix processes* worked on extending the functionality of Amdocs' 'Rater' product for two successive versions, writing shell scripts and C++ daemons* travelled to Israel for testing, and to Russia for production support of 'Rater' deployment