Tamara Makarova Data Scientist
No reviews yet

With a background that spans over 17 years in the industry, I've learned the value of real-world experience. I'm passionate about applying what I learn in real-life scenarios, understanding the theory is one thing, but I'm all about putting that knowledge to work where it really counts – in the industry.

I generally prefer project-based approach - learning by doing. It's not just about learning concepts in a classroom; it's about preparing you for the real world. I use my industry experience to show how these lessons apply outside, making learning more relevant and exciting. I believe that when you see how what you're learning today can be used in your job tomorrow, it all starts to make sense.

That's the kind of learning I stand for – practical, applicable, and always focused on preparing you for the world beyond the classroom.

Subjects

  • Statistics Beginner-Intermediate

  • Machine Learning Beginner-Intermediate

  • Data Science Beginner-Intermediate

  • Pandas Beginner-Expert

  • Python for Machine Learning Beginner-Expert

Experience

  • Senior Data Scientist (Jan, 2022–Present) at Second Foundation, Prague, Czech Republic
  • Data Scientist (May, 2020–Dec, 2021) at TickUp, Stockholm, Sweden
  • Data Scientist (Feb, 2018–Sep, 2019) at Quoine (Tokyo, Japan)
    Technologies: python (pandas, scipy, scikit-learn, keras, bokeh), ipython, SQL, bash

    * Data wrangling, cleaning and audit of customer, transaction and trading data, detection of data errors and inconsistencies, coordination of a few major data fixes
    * Analysis of high-frequency trading data for different crypto markets (order book data flows); time series models, prediction models for volatility and other financial metrics ( linear models, ensemble tree models); online algorithms and RNN, optimization and backtesting of market-making strategies
    * Design and development of flexible R&D optimization and backtesting platform in python. Coordination of the development team for the deployment of new models and features in production.
  • Data Analyst Contractor (Mar, 2017–Jul, 2017) at Soft Retail (Prague, Czech Republic)
    Technologies: python(pandas, bokeh, scikit-learn), PostgreSql

    * Analysis of retail and customer data; thorough data cleaning and audit, unsupervised anomaly detection algorithms (Local outlier factor, Isolation forest, One-Class SVM).
    * Development of effective indices and indicators for analysis and visualization of profit trends, customer segments and customer behavior trends. Developed interactive online dashboards to track main performance indices and work more closely with different customer trends and segments.
  • Quantitative Analyst (Jul, 2009–Dec, 2016) at Applied Technologies (Russia / Czech Republic)
    Technologies: Java, SQL, Excel, STATISTICA, MatLab

    * Development of various stock trading strategies based on historical financial data coming from different resources: SEC13F reports, historical stock prices and other financial reports.
    * Analysis of different data extracts, data cleaning, feature engineering, regression analysis and statistical inference (linear regression, significance tests, multiple testing problems), decision trees, clustering (K-means, hierarchical clustering). Using ideas of Markowitz efficient portfolio, Capital asset pricing model (CAPM), Jensen's alpha.
    * Backtesting, stress testing and thorough analysis of trading strategies performance. Active communication with investors and managers, presentation and reporting. Management of live trading platform.
  • Lecture Assistant (Jul, 2007–May, 2010) at Chelyabinsk State University
    * Organization of practical and laboratory classes (Econometrics, Probability Theory and Statistics, Monte-Carlo techniques).

    Involvement in a few research projects devoted to statistics and mathematical modeling:
    * Scientific project devoted to evaluation of parameter estimates for linear regression when both dependent and independent variables are measured with errors. An estimation method and statistical test for the identification of errors in both variables were developed for simple and multivariate regression.
    * Scientific project focused on the development of an accurate and noninvasive method for determination of optical parameters of biological tissues which can be further used in medical laser treatments. The developed algorithm was based on Monte Carlo simulation of a set of specific random Markov processes which due to proposed unique transformation were possible to simulate as dependent and therefore significantly decrease variance and get more accurate final results. The developed algorithm was published and then further used in the manufacture of the real medical device.

Education

  • Master degree in Applied Mathematics (Sep, 2007–Jul, 2009) from Chelyabinsk State University
  • Bachelor degree in Applied Mathematics (Sep, 2003–Jul, 2007) from Chelyabinsk State University

Fee details

    Kč500–1,000/hour (US$21–42/hour)

Reviews

No reviews yet. Be the first one to review this tutor.