For world-leading publishers like The New York Times, Chicago Tribune, USA Today, LA Times, Chronicle Herald, and others, we deliver a custom engagement platform consisted of several modules. Each module solves a specific problem in the chain of business management based on a new user, ending with his disconnection from the subscription and building analytics.The major role is played by the data we receive from the customer system. This data in our system is processed, managed, supplemented by analytics, and synchronized back.
The project is dynamic, voluminous and requires skills to work in a changing environment. There is an opportunity to study and apply automation - mix QA or AQA. The product has a large number of integrations with different authorization and circulation systems, payment providers.
We are integrated with systems such as:
- Vindicia CashBox;
- Spreedly, Braintree, Cybersource, Edgil;
- Auth0, Janrain;
- Google (SWG, Google Pay, Google Play, AMP, GA);
- Apple pay, iTunes.
About the project:
1 Data Scientist / 1 Golang Dev
What we expect:
- common knowledges: Machine Learning, Deep Learning, NLP, recommender systems, data manipulations with big data (preferably in GCP), work with DWH;
- OS: knowledges of Linux user, service manipulations, BASH scripting;
- knowledges and skills in SQL and NoSQL (preferably PostgreSQL, BigQuery, Redis, MongoDB);
- development: Python, scikit-learn Tensorflow, Keras, Pandas, NumPy, XGBoost, CatBoost, deploy models to prod with RESTful API for Flask;
- English level: Intermediate or higher.
What you will do:
- building models based on classical algorithms such as KNN, regression, decision trees etc.;
- building and finetuning of neural networks models;
- deployment models to production;
- A/B testing, data analysis, interpretation and visualization;
- research & reporting for business needs;
- data quality control;
- active participation in the development of solutions: creating and implementation of businesslogic, participation in architecture design of ETL-pipelines and DWH;
- communication with the customer, understanding of the customer’s problems from a business perspective, decomposition of the problem/requirements for subtasks;
- knowledge sharing and internal trainings.
Nice to have:
- familiar with Docker, Jenkins (simple jobs and pipelines);
- familiar with visualization & dashboarding using Google Data Studio;
- familiar with Terraform;
- have Golang knowledge/experience;
- have experience in Kaggle competitions;
- have certificates from Yandex or Google courses in Coursera (DS or Big Data Specializations);
- familiar with ETL Tools: Spark, Apache Beam;
- familiar with Kubeflow or other reproducible model deployment tools;
- knowledges in Google Cloud Platform: common architecture understanding, cloud load balancing, managedgroups, GCS, GCE, PubSub, Cloud Dataflow, Dataproc, BigQuery, Cloud Source repositories (Git), Cloud Build & triggers, AI Platform for model deploy;
- postgraduate student (have researcher’s diploma).