• Created a production bulk processing system for importing marketing assets and customer conversion data, utilizing a JSON configuration file for generic flat-file processing.
• Utilized the Python CSV library to allow a file with an arbitrary number of columns in any order to be parsed.
• Implemented string token processor to allow for an arbitrary number of fields from the input file to be combined in a single field of the output file.
• Utilized by travel industry client to create 500k+ custom ads where the text displayed was populated by replacing product-name tokens with their native-language equivalents.
• Allowed a travel industry client to be included in a beta-release of Nanigans’ real-time bidder.
• Implemented data filtering logic that allowed any function from the Python standard library to be specified for evaluating what input data should be ingested into Nanigans’ production system.
• Allowed large travel-industry client to utilize data exported from their revenue-attribution system.
• Extended configuration file spec to allow for field-specific regular expressions to be defined for extracting data.
• Utilized by clients who generated brick-mortar sales via online ads using data created by POS systems.
• Created an automated test suite (Python unittest library) of all known file processing use-cases.
• Created Oozie workflow in Hadoop to process billions of rows of raw real-time bidder data.
• Collaborated with clients’ data scientist to interpret calculated results of the A/B test.
• Acted as the lead technical contact for an ad-performance measurement project for Nanigans’ largest client.
• Created Oozie workflow in Hadoop to process billions of rows of raw real-time bidder data.
• Collaborated with clients’ data scientist to interpret calculated results of the A/B test.
• Created custom web portal (JavaScript) that allowed non-technical users to create ads based on a new specification.
• Allowed Nanigans to evaluate supporting new ad formats without committing to full productization.

• Developed a data-driven application for determining the eligibility of potential market research participants.
• Optimized the SQL Server database indexes based on typical use patterns to reduce analysis time by 75%.

• Automated internal data-processing for the customer-facing sales presentations.
• Created an automated report for calculating the bonuses for 100+ individuals.

• Languages: JavaScript/ES2015 (advanced), Python (advanced), PHP (proficient), SQL (MySQL and T-SQL), HiveQL
• Web technologies: MEAN Stack (MongoDB, AngularJS, Express.js, Node.js), Hadoop, Socket.io, Facebook Graph API
• Development Tools: Grunt, Yoeman, Bower, GitHub, VirtualBox, Chrome Dev Tools, Sublime Text
• Analytical Technologies: D3, R, NumPy, Excel, AMPL (linear programming)

• Live Demo: lorenbarcus.com/dj-demo
• Created a single page application using AngularJS where users could search for songs (via the SoundCloud API), play live music, chat with other users via web sockets (Socket.io), and provide real-time feedback to the current DJ.
• Implemented RESTful web API using Node.js and Express.js with data persisted to MongoDB.
• Developed real-time social interaction functionality allowing members of the same station to communicate.
• Implemented logic to ensure a users’ web socket was associated to a specific station upon joining.
• Ensured communication stream for the current song was re-broadcast to a new user upon joining a station.
• Utilized mongoose (MongoDB NodeJS library) define object multiple schemas and define relationships.

• Created a single page application using AngularJS where users could search for songs (via the SoundCloud API), play live music, chat with other users via web sockets (Socket.io), and provide real-time feedback to the current DJ.
• Implemented RESTful web API using Node.js and Express.js with data persisted to MongoDB.
• Developed real-time social interaction functionality allowing members of the same station to communicate.
• Implemented logic to ensure a users’ web socket was associated to a specific station upon joining.
• Ensured communication stream for the current song was re-broadcast to a new user upon joining a station.
• Utilized mongoose (MongoDB NodeJS library) define object multiple schemas and define relationships.

Senior Project: An Integer Programming Approach to Optimizing the Assignment of Professors to Classes and Timeslots.

• MongoDB University - M101JS: MongoDB for Node.js Developers (March 2016)
• Certificate: lorenbarcus.com/mdc
• Final Project: Implemented all MongoDB query documents for a demo ecommerce site and added the indexes necessary to efficiently retrieve results. Live Demo: loren.barcus.com/mm-demo
• University of Illinois Python Programming Certificate (4 courses – completed April 2013)