Summary For over a decade, one person held the belief that systematic trading could be made more lucrative, safer, and more innovative. Applying math, engineering, and computer science skills to solve the market. Financial markets are getting more complex, driven by competitive pressure. This makes the Quant-work interesting and pushes to improve skills in finding unexpected solutions. Dmitry produced creative solutions through the latest advancements in Artificial Intelligence and Machine Learning to deal with the rising complexity of practical financial market challenges. That is why major financial companies hired Dmitry to find priceless answers and bring the solutions business needed. Ph.D. in Mathematical and Quantitative Methods in Economics Greencard holder (by EB-1A/E11 Visa) |
SKILLS
- Programming:
- C / C++
- C#
- Python (Numpy, Pandas, Numba, scikit-learn, PyTorch, matplotlib).
- Solution Architecture Design.
- Data structures, Data Storing / Distribution / Mining / Management systems.
- Event-driven systems, Distributed Calculation large-scale multiserver systems, Rest API, Modular component systems.
- User interfaces, API Interfaces, and Reporting generation systems.
- Live-trading systems, System integrity checking protocols, emergency reserve systems deployment, and signaling design.
- Calculation performance optimization.
- Analytical:
- Deep Learning (ANN, Convulsional NN, LSTM, Autoencoders, GAN, VAE, Transformers)
- Reinforcement Learning (Deep Q-Learning, A3C/A2C, TDDDPG, PPO, Policy Gradients)
- Machine Learning (LightGBM, XGBoost, SVM, PCA, Random forest, Cluster Analyses)
- Genetic Algorithms / Evolution Algorithms
- Nonconvex, multi constraints, numeric optimization
- Computational Neuroscience:
- Universal plasticity rule (for flexible connectivity topology),
- Sparsely connected, spiking, Neurons/Interneurons computations circuits,
- Microcolumns structure & non-gradient-decent learning algorithms,
- Layers functionality / Connectivity, Basal Ganaglia, Thalamus, Hippocampus computational Insights in AI/Ml.
- Financial:
- Systematic trading. High/Med/Low-frequency trading (1min. - few months)
- Equity, FX, CFD, Indices, Commodities, Futures, ETF, Crypto
- Statistical and Mathematical analysis.
- Momentum, Trend following, Stat-arb, Alphas, Signals, ML/AI trading.
- Technical Analyse
- Portfolio optimization ( of assets, alphas, factors, systematic strategies)
- Financial Time Series Forecasting (ARMA, GARCH, Statistical, Markov models, ML/AI/GAN )
- Risk analysis (VAR, CVAR, PCA, Robustness test, Overoptimization checks, Labels Risk scenarios)
- Automated Alphas, Trading Strategies, and market Signals generation systems.
- Alternative Data Mining, Analyse, practical use (including NLP)
SELECTED ACHIEVEMENTS:
- Brain-Inspired new AI techniquesDriven by the curiosity to learn further techniques in solving challenging AI / ML / Quant problems, Dmitry was encouraged to find insights from Computational Neuroscience research. As a result, found organizing and functional principles of various Brain circuits and, inspired by it, introduced a set of novel AI techniques.
( Read More ) - Systematic Strategies GeneratorPresented a vision and implemented the "Quantitative Machine" for the automated generation of systematic trading strategies at SHUAA Capital. A notable financial institution acquired the technological solution in Abu Dhabi, UAE, in 2021.
( Read More ) - Advanced Analytical SolutionThe intelligent analytical system became a new standard at FxPro. Predicted EURCHF Flash Crash on Jan 15, 2015. The day when many major and oldest brokers and financial companies became bankrupt, while FxPro was prepared and remained operational through the crisis. ( How Forex Brokers Went Bankrupt overnight amid EURCHF Flash Crash Infographic )
WORK EXPERIENCE
June 2021 - Jan 2023
June 2021 - Jan 2023
QUANTITATIVE / ML / AI RESEARCHER INSPIRED BY COMPUTATIONAL NEUROSCIENTIST
Non-Compete period up to the end of 2022, Florida, USA - As a Data Researcher, I mimic Computational Neuroscience hypotheses and apply scientific / engineering methods to present new solutions for cracking challenging AI/ML/Quant tasks.
- Aiming to solve diverse Quantitative tasks focusing on delivering advanced algorithmic trading strategy for financial markets. Including data mining, asset allocation, risk management, robustness validation, etc.
- Presented a prototype of Brain-Like AI with the key features of fast learning, resistance to overoptimization, computational efficiency, supporting huge models, continuous learning, reinforcement learning, and others.
- Powered by a novel developed bio-neuron and a model of cortical microcolumn as a basic computation unit allowed solving some of the complicated problems of ML/AI easily. The flexible and changing topology of bio-neural networks, sparse connectivity, and vastly parallel computation are a few notable advancements.
- Explored a simple and universal learning rule for bio-neuron, updating weights and mechanism of growing or running dendrites. Moreover, the Brain-Like learning rule for network parts was discovered using L1 of microcolumn and apical dendrites.
- Model of artificial Basal Ganglia (for decision making), Hippocampus (data coding and storing), and Thalamus (information flow management), as a core part of novel AI system, which makes it possible to solve very challenging tasks in a completely new way.
( Read More )
Apr 2018 - May 2021
Apr 2018 - May 2021
SENIOR QUANTITATIVE RESEARCHER
SHUAA Capital, Abu Dhabi, United Arab Emirates - [AUM ~ $14B+] - Leading and Oldest (founded 1979) Asset Management and Investment Banking Company.
- Presented a vision of a Quantitative Machine with an extendable suite of modules for researching, developing, and implementing diverse trading strategies. The modular approach is allowed to be advanced by utilizing novel AI/ML/Quant approaches and extending functionality.
- The final solution can automatically generate systematic strategies, optimize trading systems, verify generated results for robustness and non-overfitting, generate synthetic trading instruments, and produce trading signals based on technical analyses and alternative data (+NLP).
- Moreover, managing the risks and constructing portfolios of systematic strategies. Powered by a developed fast and accurate backtesting engine, multiserver distributed computation modules, data management component, and live trading execution.
- Generated market-neutral Long/Short, always fully invested US equity conservative portfolio strategy bypassing COVID market shock with an all-time max drawdown of 0.9%, generating 31% annual profit. (Feb 2019 - May 2021)
- Stat-arb, portfolio Equity strategy, Intraday Equity CFD strategy, Med-frequency ETF trading, High/Med frequency futures, CFD, FX trading with 30%-70% annual profit and risk levels at 10-15%, Calmar ratios: 3.0 – 5.0
- SHUAA Capital CEO was impressed with the live-trading robust results and established a new legal entity in 2019 with Dmitry as the head of Quants for the ongoing development of a Quantitative trading systems generator machine.
- In 2021 "The Quant Machine" company was acquired for its technologies by a notable UAE financial institution.
Jan 2017 - Oct_2017
Jan 2017 - Oct_2017
SENIOR QUANTITATIVE TRADING ANALYST
Collective2.com, Remote - Leading US P2P trading platform, that connects investors with traders.
- Developed semi-automated tools for trading system research leveraging the latest advances in Artificial Intelligence and Machine Learning technologies.
- Proposed solutions to identify changes in market conditions when trading strategies underperform and utilized AI/ML algorithms to facilitate improved performance.
Aug 2013 – Nov 2016
Aug 2013 – Nov 2016
CHIEF QUANTITATIVE ANALYST
FxPro.com, Limassol, Cyprus – World Leading Retail CFD/FX Broker
- Pioneered an analytical environment, which became a new standard in FxPro. 70+ in-house analytical modules, API, and easy visual editing for customizing report templates. Advanced analyses of clients' training and systematic strategies revealed valuable business insights.
- Deployed solution for massive analysis of client trading activity to identify market shift signals and abnormalities in advance. Risk and Dealing departments could be prepared days ahead. For example, the system alerts substantial anomalies on EURCHF weeks before the SNB 2015 crush event. The company managed to hedge the risks and warned clients. FxPro suspended CHF trading and remained operational, while many oldest and major companies worldwide went bankrupt that day. ( How Forex Brokers Went Bankrupt Overnight amid EURCHF Flash Crash - Infographic )
- HFT/MFT trading on FX, Futures, Equity CFD, and Commodities. Trading based on ML/TA signals and In-House Forecasting Engine.
- Examined 7.3 million clients' trading reports. Uncovered precious business insights led to delivering dealing system modifications. Consequently, the Dealing department revenue increased 36% from $88 RPM to $120 RPM (Revenue Per Million volume)
- Engineered software for in-depth analyses of trading system reports providing suggestions and trading ideas to maximize customers' lifecycle and trade efficiency. Determine trading style and strategy and accordingly calculate specialized parameters. Generate trading forecasts to identify Risk probabilities, hidden risk scenarios, and appropriate enhancement ideas to provide the final report. Used as a core part of award-winning products. (Awards: "Best Automated Trading Platform 2013" from International Finance Magazine Award; and "Best Forex Product of the Year 2014" from UK Forex Awards)
Oct 2012 – Aug 2013
Oct 2012 – Aug 2013
SENIOR QUANTITATIVE RESEARCHER
Life Financial Group, Moscow, Russia - AUM ~$600M USD
- Designed and developed 20+ different Med-frequency FX/CFD/Futures strategies. 37% PnL with max risk at 5% achieved.
Jan 2011 – Jul 2012
Jan 2011 – Jul 2012
QUANTITATIVE PORTFOLIO MANAGER
FEx-Cor, Moscow, Russia - Private Asset Management Company. AUM ~$25M USD. Acquired by Investment Company in July 2012.
- Developed systematic-trading strategies portfolio. Become TOP out of 1000+ trading strategies worldwide at the "Currensee" FX/CFD P2P trading platform. Generated profit of portfolio (JUL2011 – JUL2012) 63% with max risk-level at 10%, managed $25M+.
EDUCATION
Oct 2010 - Sep 2013
Oct 2010 - Sep 2013
PhD, MATHEMATICAL AND QUANTITATIVE METHODS IN ECONOMIC SCIENCES - ABD
Tomsk State University of Control Systems and Radioelectronics, Tomsk, Russia - www.tusur.ru/en
Sep 2004 - Aug 2010
Sep 2004 - Aug 2010
DIPLOMA ( MASTER equivalent ) OF COMPUTER SCIENCE - GPA: 4.6 / 5.0
Tomsk State University of Control Systems and Radioelectronics, Tomsk, Russia - www.tusur.ru/en
- Winner of Russian Business analyses competition BIT from 700 companies in a country and one of two selected by Microsoft and Intel to represent Russia worldwide in California, USA. (Became the top 20 projects worldwide)
- Awarded by the US Civilian Research & Development Foundation as a "Best Business Model." (Project: Simple Trade Solutions (sTs), tools for trading)
Certifications
Click the link to open certificate
- Cutting-Edge AI: Deep Reinforcement Learning in Python [ Udemy ]
- Advanced AI: Deep Reinforcement Learning in Python [ Udemy ]
- Modern Reinforcement Learning: Actor-Critic Algorithms [ Udemy ]
- Deep Reinforcement Learning 2.0 [ Udemy ]
- Design Patterns in Python [ Udemy ]
- Deep Learning with Tensor Flow [ IBM Cognitive Class ]
- Cluster Analysis in Data Mining [ Coursera ]
- Pattern Discovery in Data Mining [ Coursera ]
- Data Analyst with Python Track [ DataCamp ]
- Data Manipulation with Python Track [ DataCamp ]
- Data Scientist with Python Track [ DataCamp ]
- Importing & Cleaning Data with Python Track [ DataCamp ]
- Machine Learning with Python Track [ DataCamp ]
- Python Programming Track [ DataCamp ]
- Python Programmer Track [ DataCamp ]