About Recoset

Recoset is a predictive analytics company focused on using "big data" in digital advertising.

Our Real-Time Data Operating System provides for rapid development and automated deployment of sophisticated data-driven applications. The OS takes care of introspecting the structure of data and extracting salient features for the models. It also provides I/O connectors into various data flows allowing for easy integration into a wide range of platforms and applications. Our vision is to quickly and efficiently build and iterate data-driven digital marketing applications on our platform using a robust set of development tools.

Uniquely, our OS can determine the economic value of a given data input or algorithm as applied to a particular problem.

About Our Technology

We use a service-oriented architecture, with a heavily optimized core system in C++ glued together with Javascript and zeroMQ. We do everything we can to avoid going to disk or involving multiple machines in a single calculation. Our machine learning is built off the open-source JML library. We use Redis to hold operational data and a combination of nginx and node.js to route and process incoming requests and data events. Data is stored via a combination of proprietary purpose-built technology and SQL for aggregated data. Our visualization framework is built on Protovis, and we use Graphite and Nagios for operational monitoring. We're big fans of open source of which we use copious amounts and contribute back wherever possible.

Ad Tech Focus

Recoset is currently focused on building and supporting a set of Ad Tech applications on our platform. This allows us to focus on solving complex, real-world problems and ensure we are taking a customer-centric approach to development.

We currently have two applications under development:

  1. Our Real Time Bidder buys display impressions in the most economically efficient manner, taking into account both price, availability and outcome. It's currently live on multiple DSP platforms.
  2. Our Real Time Data Scoring Engine seamlessly combines a marketer's 1st party data, with 2nd and 3rd party data to predict a probability score for each outcome for each unique user. These scores are then applied across digital channels like DSPs, PPC Platforms, E-Mail Platforms and Creative/Content Optimizers.

We see these applications as converging into one, with ever more sophisticated buying decisions being made in real-time based upon an ever-increasing amount of information.

Our goal is to become the intelligence layer in an advertiser's technology stack, enabling a truly user-centric approach to marketing. One which is channel agnostic and focuses on the right message, through the right engagement channel, at the right time.