Data Scientist

Role & responsibilities

In Data Science we work with the data from fast changing consumer goods world. Data comes both from standardized databases and from online retail channels in various forms like semi-structured text in various languages, images or precise numeric values. We have already developed 400+ ML models which help us to identify consumer good attributes our clients are interested in. Currently we are looking for a new colleague to join the team and assist developing new and improving existing ML models according to our client feedback and suggest new ways looking at the data which would result in new product

  Formulating hypotheses and developing proof-of-concept ML models for NLP, Image Processing,

•       Supervised/Unsupervised Learning tasks;

•       Translating Industry specific knowledge into proper models;

•       Monitoring and ensuring data quality;

•       Communicating insights effectively in both written and visual way;

Preferred candidate profile

We expect you to have:

•       Background in Mathematics/Physics/Natural Sciences or Engineering fields, relevant working experience of 2+ years;

•       Advanced knowledge of Python/R/SQL, understanding of merge-request git flow, basic scripting including CI/CD, regular expressions, shell commands, HTML/JavaScript and basic webapp development skills.

•       Good coding habits such as proper documentation, linting, styling, reproducibility, code review best practices, and test-driven development.

•       Skills to maintain ML models: measure performance, add/remove classes, create effective labelling batches, retrain with assessment of the impact.

•       Ability to suggest model improvements, propose working solutions for business problems, ensure the model works with existing models, track model experiments and keep model documentation updated.

•       Skills to clearly communicate problem status, actions taken, conclusions, improvements and limitations.

•       Experience creating and using cloud resources, familiarity with notebooks, SQL instances, buckets, secrets, VMs, and service accounts, understanding and contributing to pipelines, applying model scaling practices and cloud cost awareness.

•       Understanding of which visualizations to use with different data types, comfortable with at least one visualization framework, and capable of creating shareable/dynamic reports/apps.

•       Proficiency in applying EDA principles, classical supervised/unsupervised learning techniques, main NLP techniques, a good understanding of deep-learning techniques and awareness of model and computational complexity.