Responsibilities:
• Investigate large datasets looking for latent relationships
• Formalize complex relationships among available data and implement predictive models of dependent variables
• Rapid prototype and test different hypothesis
• Use Natural Language Processing techniques along with classical and novel Machine Learning methods to leverage multidimensional employee engagement and survey data
• Research, brainstorm and productize analytical predictive models
• Use of great communication skills for effective coordination with supervisor and across peers, handling project deliverables in an agile fashion
Skills and Qualifications:
• Business acumen and good intuition for effective data exploratory analysis
• Natural ability of presenting results in a clear manner to technical and non-technical audience
• Ability of suggesting and extending company’s data with third party sources of critical data, useful to increase model performance
• Use of different type of correlation analysis (Pearson, Spearman). Ability to visualize multidimensional feature relations
• Suitability and usage of statistical tests, as T-test, Chi Squared, ANOVA
• Proficiency in querying languages, using traditional RDBMS and No-Sql databases. Hands on knowledge of Snowflake.
• Strong command on data preprocessing steps, including imputation, handling of unbalanced classes, handling of collinearity. Use of different sampling techniques
• Ability to perform efficient feature engineering, feature selection, dimensionality reduction
• Proficiency in regression models and binary and multiclass classification models. Excellent understanding of classical linear and non-linear models. Experience with Decision Forests, Gradient boosted ensemble methods and Support Vector Machines. Use of K-NN and Bayesian networks
• Proficiency in using regularization methods, hyper-parameter tuning, handling of model complexity
• Proficiency with Natural Language Processing techniques. Raw text preparation and cleansing. Stemming and Lemmatization. Ability to perform Topic modeling and sentiment analysis. Experience with aspect-based opinion mining
• Good experience with unsupervised learning. Use of clustering methods. Ability to visualize n-dimensional spatial clusters. Use of K-means
• Proficiency with ANNs and deep learning. Strong knowledge of network background function and error propagation. Use of Keras with custom loss functions. Network optimization through GridSearch and/or Genetic Algorithms.
• Experience with anomaly detection. Use of Autoencoders.
• Understanding of the different modeling performance validation and model generalization. Building and interpretation of learning curves
• Ability of model interpretation, explaining feature importance for the various modeling techniques
• Good scripting and coding skills for rapid prototyping. Excellent knowledge of Python, pandas, scikit-learn and keras packages
• Knowledge of the main convex optimization techniques
• Experience with data visualization tools, such as matplotlib, seaborn and D3.js
Educational and Professional background:
• M.S. from an analytical major, such as Engineering, Statistics, Computer science, Operations research
• Minimum 3 years working experience, in a related field, such as Adtech, AI, System automation, Computational linguistic, Inventory optimization
About NCube
NCube is an English outstaffing and outsourcing company with global development centers.
Working directly with teams of professionals from England, the United States, Europe, and Canada you will certainly be able to improve your knowledge and skills.
Our main value is yours and our professionalism.
We are not a huge corporation that means a minimum of bureaucracy and maximum attention. In an atmosphere of creativity and clear management, we will try to make your work as comfortable as interesting.
Let’s rock this world!
Company website:
https://ncube.com/
DOU company page:
https://jobs.dou.ua/companies/ncube/
The job ad is no longer active
Look at the current
jobs
Data Science
Kyiv→