Browse profiles Post a project
Malt welcom

Welcome to Sergey's freelance profile!

Malt gives you access to the best freelancers for your projects. Contact Sergey to discuss your project or search for other freelancer profiles on Malt.

Sergey Mastitsky

Experienced Data Scientist

Works remotely from City of London

  • 51.5073
  • -0.127647
  • Suggested rate €720 / day
  • Experience 7+ years
Propose a project The project will begin once you accept Sergey's quote.

This freelancer is available part-time (Evenings & weekends) but hasn't confirmed their availability in over 7 days.

Part-time, Evenings & weekends

Propose a project The project will begin once you accept Sergey's quote.

Location and workplace preferences

Location
City of London, United Kingdom
Remote only
Primarily works remotely

Preferences

Project length
Would prefer:
  • ≤ 1 week
  • ≤ 1 month
Would prefer to avoid:
  • Between 3-6 months
  • ≥ 6 months
Business sector
  • Arts & Crafts
  • Automobile
  • Banking & Insurance
  • Fashion & Cosmetics
  • Defense & Military
+21 autres

Verifications

Freelancer code of conduct signed
Read the Malt code of conduct

Verified email

Reputation

Languages

  • English

    Fluent

  • Russian

    Native or bilingual

  • Belarusian

    Native or bilingual

  • German

    Basic

Categories

Skills (22)

Sergey in a few words

Trained to a PhD level, I have 20 years of experience in academic and industrial
sectors, where I used Statistics and cutting-edge Machine Learning to create new knowledge and deliver measurable business outcomes. My colleagues have repeatedly acknowledged my expertise in a wide range of modern data analysis and manipulation tools (e.g. R, Python and SQL), ability to clearly explain technical concepts to a layman, high attention to detail, friendliness and professional manner. I've also authored 40+ peer-reviewed scientific articles (including a paper in Nature) and 5 books on data analysis and visualisation.

As a Data Science consultant and freelancer, I provider a variety of services, including but not limited to:
- Deep and actionable, data-driven insights into your operational processes, customers, and products;
- Accurate models for customer lifetime value and churn to power value-based
acquisition and retention campaigns;
- Predictive applications to perform planning and optimisation tasks (revenue forecasting, budget allocation, recommender systems, preventive maintenance, etc.);
- Bespoke software development for data analysis and visualisation (e.g. interactive Shiny applications, APIs, etc.);
- Personnel training and upskilling in Data Science (including introductory sessions for executives and managers, and one-to-one coaching sessions).

Experience

Belarusian State University

Research

Assistant Professor

Minsk, Belarus

September 2004 - July 2008 (3 years and 10 months)

Responsibilities:
- Teach Statistics and other relevant subjects to Biology students
- Conduct independent research work
- Secure funding via grant applications
- Publish research results and present them and professional meetings

Key achievements:
- Developed and delivered the following courses: “Biostatistics”, “Population Ecology”, “Bioassessment of the Quality of Environment” and “Ecological Modelling”
- Managed a group of 4 researchers and provided funding for them
- Published multiple papers in peer-reviewed academic journals
- Made presentations at numerous professional meetings and public events
lecturing research grant proposal writing paper writing teaching statistical analysis statistics ecological modelling population ecology

Great Lakes Center at Buffalo State College

Research

Research Scientist

Buffalo, United States of America

August 2008 - October 2009 (1 year and 2 months)

Responsibilities:
- Design and implement research projects to assess ecological impacts cased by invasive species in the Great Lakes region
- Conduct field sampling and laboratory sample processing, followed by data analysis
- Secure funding via grant applications
- Publish findings and present them at professional meetings

Key achievements:
- Research results published in a number of high-impact peer-reviewed academic journals
- Presentations made at multiple professional meetings and public events
grant proposal writing field work microscopy statistical analysis paper writing

German Cancer Research Center

Research

Postdoctoral Researcher

Heidelberg, Germany

June 2010 - March 2012 (1 year and 9 months)

Responsibilities:
- Analyse large multivariate datasets generated in genome-wide cancer studies in order to identify differentially expressed genes and biomarkers indicative of the disease progression
- Develop pipelines for statistical analysis of gene expression data obtained in longitudinal studies
- Publish findings and present them at professional meetings

Key achievements:
- Research results published in a number of academic papers, including a publication in Nature
- Presentations made at multiple professional meetings

BASF

Chemical

Statistical Modeller

Limburgerhof, Germany

April 2012 - October 2014 (2 years and 6 months)

Responsibilities:
- Conduct routine statistical analyses of data obtained in field and laboratory trials
- Model the effects of agrochemicals on populations of wildlife
- Develop and deploy data products
- Write up statistical parts of the product dossiers
- Provide in-house training on statistical methodologies

Key achievements:
- Performed complex statistical modeling and simulations related to the assessment of environmental risks posed by agrochemicals (e.g., fungicides)
- Developed a number of web applications (e.g., for dose-response modelling) that facilitate the analysis and report generation in laboratory studies
- Developed a individual-based population model for assessment of acute toxicological risks posed by one of the products to small mammals
- Prepared multiple internal research papers, as we as reports submitted to regulatory bodies across the globe
- A paper on the use of individual-based population models for ecological risk assessment published
in a peer-reviewed academic journal
- Multiple training sessions on Statistics and R programming

Technologies used in the above projects: R, RStudio, Shiny, NetLogo, PRIMER-E, GraphPad Prism, Git, LaTeX
R RStudio Shiny NetLogo PRIMER-E GraphPad Prism Git LaTeX

RNT Consulting

Consulting & Audits

Data Analyst  - As a freelancer

Heidelberg, Germany

October 2010 - July 2014 (3 years and 9 months)

Responsibilities:
- Consult on the design of field and laboratory experiments
- Analyse experimental data
- Write up statistical parts of the reports for clients

Key achievements:
- Design and statistical analysis of experiments conducted by RNT Consulting (www.rntconsulting.net) for their industrial clients
- Project reports prepared for such major clients as the U.S. Bureau of Reclamation, California Department of Water Resources, and Central Arizona Water Conservation District
- Two papers published in peer-reviewed academic journals
R RStudio STATISTICA design of experiment field experiments

Millicom

Telecommunications

Data Scientist

City of London, United Kingdom

August 2014 - October 2015 (1 year and 2 months)

Responsibilities:
- Work closely with stakeholders from different functional units (Customer Understanding, Marketing,
Pricing) and provide Data Science solutions to their business problems
- Develop and deploy data products
- Provide in-house training on Data Science best practices and programming

Key achievements:
- Developed a scoring model for zero-balance products. The model accurately identifies risky customers and minimizes the probability of default by recommending optimal loan amounts to non-risky customers
- Developed and deployed customer churn models for 3 different operations in Africa
- Develped a customer segmentation methodology based on the customers' behaviour and expected value. This methodology was used to develop new products relevant for the respective segments
- Used social network analysis to identify influential customers and then recruit them for viral marketing campaigns
- Developed and deployed a web application for estimating the impact of below-the-line and above-the-line market interventions
R RStudio Shiny H2O R Markdown SQL Git LaTeX

Derivco

Software Publishing

Lead Data Scientist

City of London, United Kingdom

November 2015 - October 2018 (2 years and 11 months)

Responsibilities:
- Identify key areas of the business where Data Science can add value
- Prototype and build data products
- Design and analyse controlled experiments and observational studies
- Manage team members
- Explore novel Data Science-related technologies that can benefit the company
- Promote the culture of data-driven decision making in the company

Key achievements:
- Recommender for casino games: a collaborative filtering-based system that ranks casino games according to their relevance to a given player, while maximizing the expected revenue
- Recommender for personalised deposit values: a system that predicts the most likely next amount to be deposited by a given player, which is then used on the UI to improve player experience and drive upselling
- Application for automated forecasting of the load on IT systems of a major sportsbook (e.g., number of customers, number of bets, etc.), with a web-based frontend
- API Monitoring Tool with an anomaly detection module: a web-based application that visualises real-time performance of multiple internal-layer APIs in a human-readable form and sends email alerts when anomalies in performance are detected
- Regular "Data Science Team Newsletter", sent out across the company in order to provide updates on the team's projects and to share knowledge
R RStudio Shiny Shiny Server R Markdown MySQL Mahout Python H2O AWS S3 AWS EC2 AWS Lambda AWS EMR AWS SFN AWS API Gateway AWS SES AWS SNS Git LaTeX

Aviva

Banking & Insurance

Data Science Lead

City of London, United Kingdom

October 2018 - Today (4 years and 1 month)

Main duties:
– Manage a team of 3 Data Scientists & 2 Data Engineers, set & reinforce the team culture
– Coach team members & ensure their constant development
– Data-driven optimisation of CRM & Marketing activities
Example projects:
– Recommender system for insurance products, delivering double-digit uplifts in sales across channels
– Automated system to infer intent from customer’s web behaviour
– Automated system to categorise Google keywords into semantic groups for paid search optimisation (presented at the conference “Enterprise Applications of the R Language”, London, 2019)
Dataiku Data Science Studio R Python Spark H2O Keras gensim fastText word2vec topic modelling Adobe Marketing Cloud RStudio SQL Git

Education