Mikhail Dubov

Mikhail Dubov

Data Scientist & Software Engineer

Work Experience

Revolut

Lead Data Scientist • Revolut Business

London • Mar 2023 – present

Revolut Business offers global business bank accounts in Europe, US, Australia and Asia.

  • Led a team of 3 data scientists that delivered ML projects for card fraud (5x block rate decrease; £100k/mo saved in chargebacks), receipt matching (migration to an in-house CV model saving $10k+/mo), and LLM-based customer feedback categorisation
  • Led BI migration from a data warehouse to a data lake backend that reduced the loading times of 200+ dashboards by 95% through data partitioning and use of aggregate tables
  • Implemented and led the quarterly KPI review process for 20+ teams in the department

Senior Data Scientist • Revolut Business

London • Jul 2021 – Feb 2023

  • Managed the central B2B data team, grew it 2x to 10 DA / DS / DE, and implemented product team embeddings, facilitating the support of 15+ product teams with analytics
  • Prototyped and passed CTO's system design review for a company-wide A/B testing framework, later adopted by 25+ teams in the company (Python / Airflow / Trino)
  • Led the migration of 60+ dashboards from Metabase to Looker (SQL / LookML)

Data Scientist • Revolut Business

London • Jul 2020 – Jun 2021

  • Designed an ML-based system for automated receipt parsing & matching for expense management, resulting in a 2x increase of the product retention rate (Python / GCP)
  • Built ETL pipelines driving the migration of Brexit-impacted businesses to a new legal entity, saving £10m+ in annual revenue that would be lost otherwise (SQL / Airflow)
  • Built a family of ML models for customer LTV; productionised them in Python & Airflow (digital marketing optimisation) / in Java for online predictions (KYC ticket prioritisation)
  • Built a system for collection, anonymisation, and statistical analysis of customer NPS & feedback, driving the product roadmap and 6 KPIs (Python / spaCy / Airflow / Looker)
  • Built a production-grade library for robust feature computation (Python / SQLAlchemy)
  • Built time series forecasting models for business signups and card spend (FB Prophet)

Smarkets

Quant Developer • Trading

London • Oct 2018 – Jul 2020

Smarkets is one of Europe's leading platforms for algorithmic sports trading.

  • Worked on 5 trading strategies for market making / prop trading on Smarkets and other exchanges. Owned one of these strategies from the initial design and implementation (Python / Docker / K8s / AWS) to maintenance and ops (£100k+/qtr, 8.5 ann. Sharpe)
  • Built ML systems for categorising trading activity and automating hedging decisions
  • Developed an internal Python library for quant data loading and transformation
  • Extended and optimised the P&L analytics system (Python / SQLAlchemy / BigQuery)

Data Scientist • Marketing

London • Dec 2016 – Sep 2018

  • Built ML models and ETL pipelines to predict customer churn & LTV, driving marketing campaigns that improved retention by 25% in A/B testing (Python / Luigi / Sklearn)
  • Built an internal anomaly detection system based on time series forecasting (FB Prophet)
  • Designed the data analyst interview process and interviewed 50+ DA / DS candidates

Mirantis

Software Engineer • OpenStack Performance

Moscow • Jan 2015 – Jun 2015

OpenStack is an open-source cloud computing platform used by CERN, Baidu, and others.

  • Was one of the core developers in OpenStack Rally (open-source benchmarking tool, 300+ GitHub stars) and top #2 contributor with 100+ commits, 900+ code reviews
  • Mentored the Rally developer community and created extensive documentation

Junior Software Engineer • OpenStack Performance

Moscow • Aug 2013 – Dec 2014

  • Implemented core components in Rally: benchmark launchers, data processing tools, API
  • Achieved a 5x speed-up in the node listing algorithm of OpenStack Nova (platform core)

Internships

Google

Software Engineering Intern • Forecasting

London • 2015

  • Built a client app for an internal time series forecasting platform (Java)
  • Added support for ML-based time series analysis (Google Prediction API)

App in the Air

Data Science Intern • Reviews

Moscow • 2012

  • Built an ML system for the classification of airport reviews

Education

Université Paris-Est Marne-la-Vallée

M. Sc. in Computer Science

Paris • Oct 2015 – Sep 2016

mention très bien

Higher School of Economics

M. Sc. in Data Science
B. Sc. in Software Engineering

Moscow • Sep 2010 – Oct 2016

GPA 9.6/10, ranked #1 in class

Linguistic Gymnasium 1513

Specialisation: German language

Moscow • Sep 2003 – Jun 2010

Gold medal, ranked #1 in class

Skills

  • Languages: Python, Java, SQL, previous experience with C#, F#, R
  • Libraries: Pandas, Sklearn, Numpy, Scipy, FB Prophet, SQLAlchemy, PyTorch
  • Software engineering: Git, Docker, Kubernetes, Pytest, JUnit, AWS, GCP
  • Data engineering: Airflow, Luigi, Postgres, Redshift, BigQuery, Exasol
  • Business intelligence: Looker / LookML, Sisense, Metabase

Languages

  • English (IELTS 8.5/9.0)
  • German (DSD C1)
  • Russian (native)