Position: Middle/Senior Data scientist

Where? Alfa-Bank, Model Development Department, Advanced Analytics Directorate for Legal Entities.

Work format: Hybrid: Moscow, Tekhnopark metro station + remote.

I am actively hiring candidates for middle/senior data scientist positions in the model development team for Alfa-Bank's legal entity clients. We are currently looking for roles in developing a recommendation engine for bank products in digital channels and assessing the potential profitability of clients for the bank (CLTV), but there are also other ambitious and interesting tasks.

What do you need to do?

1. Analytics of semi-structured data in Hadoop, creating new features and applying them in models.
2. Experimenting with approaches non-standard for classical scoring for modeling.
3. Analysis of the effectiveness of new data sources, testing credit risk models on new data sources.
4. Full cycle of development and productionization of classic ML models: — participate in regular video calls with business customers. Formulating the basic technical specifications jointly with the customer; — conducting primary analytics, demonstrating results to the customer, adjusting requirements; — managing customer expectations regarding the final result and expected timelines for development, testing, and implementation of the solution; — developing ML model(s) to solve a business task from scratch, i.e., forming a general dataset and target event using SQL queries in the database, developing one or several ML models for the most optimal solution of the task, developing additional features to improve prediction quality if necessary.
5. Preparing materials (presentations) to demonstrate work results.
6. Productionization of the model solution.

Bonus: A friendly professional team of more than 30 data scientists working in various directions to develop processes related to the corporate sector of the economy.

Requirements: — Education in physics and mathematics, economics and mathematics. Good knowledge of probability theory and mathematical statistics; — Ability to work with data (analysis, cleaning, preparation, feature selection and creation), skills in writing SQL queries. Experience working with Hadoop (Hive, Impala, Apache Spark). — Knowledge of Python (specifically, libraries used for data analysis: SciPy, pandas, scikit-learn, XGBoost, LightGBM, etc.); — Experience in developing classical ML models (linear models, trees, boosting, etc.). Knowledge of data processing methods (preparing datasets for training), quality assessment metrics. Understanding of the principles of operation of classical ML model algorithms. — Communication skills: Required skills (and experience) in communicating with business customers, i.e., a person or group of people without ML competencies and deep analytics competencies, but who well understand the process needs for enhancement with ML models.