Research Scientist

About Media.net

Media.net a Directi company, powers the Yahoo! Bing Network Contextual Ads program and is a leading global contextual advertising company that provides a wide range of advertising and traffic monetization solutions. Our diverse client base includes web publishers with premium content as well as registrars, hosting companies and holders of high value domain portfolios. We aim to consistently deliver long-term superior performance by using leading-edge technology, high-quality research, and proprietary processes that allow us to display the most precisely targeted, highest paying sponsored links and ad topics.

We serve our contextual real-time ads customized for each visitor and each page view across billions of visitors, across 10’s of millions sites/domains on a server infrastructure that runs 1000+ CPUs, terabytes of RAM, and 1000’s of terabytes of data.

What is the job like?

This team is responsible for building systems that include massive database infrastructure, relevance algorithms, ad delivery engines, data mining systems, web user interfaces etc. The problem space encompasses some of the most comprehensive problems in Computer Science (Machine Learning, Data Mining, BI, Large Scale Distributed Computing and Artificial Intelligence); Mathematics; Economics & Business (Auction Theory, Behavioral Targeting).

The team’s goals is to build algorithms to map the user’s intent with the content of the ads that are shown. The team is directly responsible for increasing company revenues.

  • On one side this is done by using Natural language Processing & Information retrieval techniques to find out what is the given page talking about. Commercial topics, categories etc from the contents of the webpage as well as the site are extracted and used to deliver the most relevant ads.
  • On the other side feedback from user actions are coupled with sophisticated statistical and Machine learning techniques to improve quality of ads over time. Aggregations are analyzed by systems over url, domain, category and topic of page to be able to better predict what is likely to do well for a given impression.
  • Another of the upcoming charters of the team is to show better ads using behavioral targeting based on specific interests of that user on a general web page.

Responsibilities include:

  • Working with engineering teams to co-develop a research agenda and evaluation metrics; Establishing priorities and the tactical plan for implementing this strategy;
  • Using research and software skills to implement a prototype and report quantitative performance metrics;
  • Working with engineering teams to productize the prototype; Participate in and contribute to technical discussions
  • You will have strong theory/algorithm background and very good understanding on how to apply advanced knowledge to solve real problems.
  • You will also have the ability to effectively create research agendas and communicate innovations to the rest of the organization.
  • Utilize advanced data mining and machine learning to personalize the user experience, increase user engagement and improve core business metrics

Who should apply for this role?

  • Experience in Artificial Intelligence or Information technology or related field
  • Should be able to solve complex algorithms
  • Knowledge of machine learning or text mining, or natural language processing
  • Excellent analytical, logical and problem solving skills
  • Excellent verbal & written communication skills
  • Ph.D or Equivalent in Computer Science, Statistics, Information Technology or related field 
  • Knowledge of Big Data and Cloud computing technologies
  • Hands on coding experience
  • Entrepreneurial thinking
  • Publications or presentation in recognized Machine Learning and Data Mining journals/conferences such as ICML
  • Knowledge in several of the following: Math/math modeling, decision theory, fuzzy logic, Bayesian techniques, optimization techniques, statistical analysis of data, information retrieval, natural language processing, large scale data processing and data mining
  • Ability to deal with ambiguity & break them down into research problems
  • Strong theoretical and research acumen

Benefits & Perks

Our excellent benefits include everything from great medical and life insurance to catered meals. Our workspaces are comfortable and fun, complete with bean bag chairs, ping pong tables, and all the snacks you can eat. We have no dress code (tee-shirts are a-ok!). We have flexible work hours and flexible holidays, which means that teams pick their own work hours. Media.Net has its own concierge desk that doubles up as a travel agency. If you like the idea of working in an exciting workspace on cutting-edge internet products that make a truly global impact (and wearing flip-flops to work), then we want to get to know you!