Alexandr Proskurin, Developer in London, United Kingdom
Alexandr is available for hire
Hire Alexandr

Alexandr Proskurin

Verified Expert  in Engineering

Bio

Alexandr is an AI developer and entrepreneur who co-founded two companies. He specializes in financial machine learning techniques applied in investment management and algorithmic trading research. His expertise includes researching and implementing long-only and intraday long/short and statistical arbitrage algorithms using supervised and unsupervised algorithms. Alexandr developed MlFinLab, a Python package for financial machine learning research used by portfolio managers and traders.

Portfolio

Principia Invest
Python, Bloomberg API, Scikit-learn, Amazon S3 (AWS S3), DigitalOcean
Hudson & Thames
Python, MlFinLab
Modex
Python, Pandas, Apache Arrow, SQL, NumPy, Scikit-learn

Experience

Statistics - 10 yearsData Science - 6 yearsPython - 6 yearsPortfolio Management - 6 yearsMachine Learning - 6 yearsAlgorithmic Trading - 5 yearsComputer Science - 5 years

Availability

Part-time

Preferred Environment

Ubuntu Linux, PyCharm, Jupyter Notebook

The most amazing...

...implementation I delivered was the intraday momentum machine learning algorithm trading SRW Wheat futures.

Work Experience

Co-founder and CIO

2020 - PRESENT
Principia Invest
  • Managed a team of two researchers and a financial analyst who have researched and implemented multi-factor long-only US equities algorithms with an unsupervised learning portfolio optimization component. (CAGR 19% since 2008, Sharpe 1.1).
  • Delivered a real-time implementation of a portfolio algorithm using Bloomberg API and Python on DigitalOcean hosted server and S3 data lake.
  • Initiated a real-time database update on 1,500+ liquid US stocks including market, fundamental, and sentiment data.
Technologies: Python, Bloomberg API, Scikit-learn, Amazon S3 (AWS S3), DigitalOcean

Co-founder and Head of Consulting

2018 - PRESENT
Hudson & Thames
  • Developed MlFinLab - a Python package for financial machine learning research. Managed at least six open-source Python developers.
  • Participated in the development of ArbitrageLab - a Python package used to conduct a research in pairs trading, mean-reversion, and statistical arbitrage. Managed the cointegration approach and Kalman filter implementations.
  • Assisted in the development of PortfolioLab - a Python package which contains various algorithms used in portfolio optimization,.
Technologies: Python, MlFinLab

Senior Quantitative Researcher

2018 - 2020
Modex
  • Researched and implemented a VIX futures trading strategy with an intraday hedging component (1-Minute Bars) (40% ROC since January 2019 and Sharpe ratio 2.2 in the 2012-2019 backtest period).
  • Managed a back-end engineer and quantitative developer who designed and implemented a high-performance proprietary backtesting platform (Apache Arrow, Parquet, Hadoop stack) with a team of software developers.
  • Improved the existing FX strategy (increased Sharpe ratio from 0.7 to 1.2 in the 2010- 2019 backtested period with 12% ROC since August 2018 in real-time trading).
Technologies: Python, Pandas, Apache Arrow, SQL, NumPy, Scikit-learn

Quantitative Researcher

2016 - 2018
Integral Capital Management Sarl
  • Implemented an index option trading strategy (13% ROC for six months of trading, backtested performance: 30% CAGR with 23% volatility for the 2010-2017 backtest period).
  • Participated in creating a proprietary API for multithreading financial data preprocessing and feature generation in Python.
  • Implemented a quantitative market ETF management strategy with monthly rebalance (CAGR 11%, Sharpe 1.0 in the 2008-2017 period).
Technologies: Python, SQL, Scikit-learn

Journal of Financial Data Science Scientific Paper

My paper "Does the CFTC report have a predictive power: Machine Learning Approach" was accepted to a Journal of Financial Data Science, summer issue 2021. In the paper, I've applied financial machine learning techniques to extract informative features from the CFTC COT report to predict the prices of SRW Wheat, Corn and Soybean futures.

MlFinlab Package

https://github.com/hudson-and-thames/mlfinlab
MlFinLab is a collection of production-ready algorithms (from the best journals and graduate-level textbooks), packed into a Python library that enables portfolio managers and traders greater insight who want to leverage the power of machine learning by providing reproducible, interpretable, and easy-to-use tools.

ArbitrageLab Python Package

https://hudsonthames.org/arbitragelab/
ArbitrageLab is a collection of algorithms from the best academic journals and graduate-level textbooks, which focuses on the branch of statistical arbitrage known as pairs trading. We have extended the implementations to include the latest methods that trade a portfolio of n-assets (mean-reverting portfolios).
I was responsible for implementations of the cointegration approach and dynamic hedge ratio estimation using Kalman filter.
2016 - 2018

Master's Degree in Banking and Finance

Kyiv-Mohyla Business School - Kyiv, Ukraine

2012 - 2016

Bachelor's Degree in System Analysis

Kyiv Polytechnic University - Kyiv, Ukraine

OCTOBER 2018 - PRESENT

Algorithmic Toolbox

Coursera

OCTOBER 2018 - PRESENT

Data Structures

Coursera

OCTOBER 2018 - PRESENT

Algorithms on Graphs

Coursera

Libraries/APIs

Pandas, NumPy, MlFinLab, Scikit-learn, SciPy, Bloomberg API

Tools

PyCharm

Languages

Python, SQL

Platforms

Jupyter Notebook, Apache Arrow, DigitalOcean, Amazon Web Services (AWS)

Storage

Amazon S3 (AWS S3), PostgreSQL

Other

Statistics, Algorithmic Trading, Portfolio Management, Asset Management, Machine Learning, Data Science, Optimization, Trading, Options Trading, Futures & Options, Mathematics, Time Series, Time Series Analysis, Quantitative Modeling, Quantitative Finance, Artificial Intelligence (AI), Computer Science, Finance, Data Engineering, Calculus, Linear Algebra, Algorithms, Investments

Collaboration That Works

How to Work with Toptal

Toptal matches you directly with global industry experts from our network in hours—not weeks or months.

1

Share your needs

Discuss your requirements and refine your scope in a call with a Toptal domain expert.
2

Choose your talent

Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.
3

Start your risk-free talent trial

Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.

Top talent is in high demand.

Start hiring