Some Recent Work

See below some of our recent work

Social TV Guide

TV Map is a Social TV platform for users to share what they are watching, see what others are watching, and connect to people who are watching the same tv shows. They have a complete TV Guide (TV Listings) for all TV stations. They also do Social Media Analytics, measuring social networks buzz, and showing in real time those TV Ratings as a guidance on what's hottest on TV.


Website was developed using ASP.NET MVC 4, using Entity Framework. Social connectors were developed for Facebook, Google Plus, and Twitter. Crawlers (developed in WPF, using Quartz.NET scheduler and TweetSharp/Hammock) are used to fetch metadata (and even some media) from multiple sources, both from EPG providers (electronic programming guides), from TV Broadcasters, from TV Networks, and from 3rd-party references (like IMDB). All data is deduplicated and passes through complex tiebreaker algorithms that determine which is the most correct information.
TV ratings for popular TV channels are measured on real-time over Twitter, through a complex heuristic-algorithm that identifies if tweets are relevant (for some specific TV show) according to their location and their message, using natural language processing (using Stanford's CoreNLP) and a neural network that learns hashtags, mentions, and frequent terms.

Lots of custom caching mechanisms were developed for scaling support. was used for content indexing and search.

Jobs for Refugees

Expatt.Org is a hub for connecting employers to expatriates or refugees.


Website was developed using ASP.NET MVC 5, using Entity Framework, Grunt/Bower. Social connectors were developed for Facebook, Google Plus, Twitter and LinkedIn. We used technologies like jQuery, Google Maps (for displaying map of where refugees are living), Google Places API (to autocomplete cities and addresses), Facebook API (for downloading user's profile picture), and jQuery AJAX plugins like select2, selectize and validation.

A complex email-masking mechanism was implemented, giving each user a virtual email-address (at that transparently forwards messages to their real address. With this infrastructure, communications between employers and candidates is totally tracked, and can be done using regular e-mail clients without revealing their real email-addresses to each other.

A Dual-language structure was developed and the user's language is automatically detected according to browser's language.