Senior Data Scientist Offline

MGID was founded in 2008 and is one of the leading companies in native advertising. We enable our media partners to monetize their audience and help brands to promote their services and goods effectively.

MGID offers a range of integrated solutions covering the promotion process every step of the way; we offer services ranging from planning out the marketing strategy to its thoughtful implementation and optimization. Our clients include major international brands like Renault, Domino’s, Airbnb, PizzaHut, Qatar Airlines, and many others, including media organizations and web agencies.

MGID is:

— One of the largest MarTech-companies in the Ukrainian market;

— A proprietary Highload service that delivers 185 billion advertisements to 850 million unique users in more than 70 languages;

— The winner of multiple AdTech awards for innovation and product quality;

— A workforce of 600+ employees operating from offices in the US, Europe, and Asia;

— A passion for cutting-edge technologies and a seamless vertical structure that allows the regional teams to exchange skills and development practices.

Responsibilities:

- Develop new models for different recommender systems within our product (classification, clusterization, etc.);

- Research, patterns identification, and formulating insights;

- Clean data from noise, find patterns for irrelevant, robot generated events;

- Identify proper evaluation metrics, make an offline and online evaluation and testing for new models;

- Upgrade and support current models;

- Enrich data with new features using first and third-party data;

- Developing applications or notebooks for maintaining models (training, usage);

- Identify new opportunities to leverage data science to different parts of the MGID product.

Primary requirements:

- 3+ years of experience, deep skills in Machine Learning\Data Science;

- Hands-on experience with machine learning concepts: regression and classification, clustering, feature selection, feature engineering, curse of dimensionality, bias-variance tradeoff, neural networks, SVMs, etc.;

- Have professional experience with Python(Scikit-learn, Pandas);

- Experience with Keras/TensorFlow

- Knowledge of Spark (Spark ML/PySpark);

- Understanding of true model validation, p-value, and equivalent;

- Good data manipulation skills are required including cleaning and managing data;

It will be a plus:

- The ideal candidate holds an MSc or Ph.D. in Engineering, Statistics, Mathematics, or related fields.

- Knowledge of R;

-Understanding of online advertising, RTB systems, know the difference between DSP, SSP, and DMP;

-Hands-on experience with NoSQL databases.

What we offer:

— Friendly team, opportunities to share your knowledge and experience;

— The newest office in the business center “Marmalade”;

— English courses with a native speaker;

— Flexible approach to the schedule;

— Corporate participation in sports, eco-and social projects.

— Health insurance package.