Adrian is available for hire
Hire AdrianAdrian Alexandru Olteanu
Verified Expert in Engineering
Quantitative Researcher and Developer
Bucharest, Romania
Toptal member since July 10, 2020
Bio
Adrian is an experienced quantitative researcher and statistician/data scientist with the ability to research, implement, and deliver predictive ML models used in finance—namely algorithmic trading. Other areas of domain expertise of Adrian's include mathematical modeling, NLP, deep learning, anomaly detection, signal processing, and portfolio optimization.
Portfolio
Two Sigma Investments
Python, Algorithmic Trading, Data Science, Artificial Intelligence (AI)...
SPS Trading
Algorithmic Trading, Python, Genetic Algorithms, Crypto, Data Science...
Tickup
Python, SQL, Algorithmic Trading, Artificial Intelligence (AI)...
Experience
Availability
Part-time
Preferred Environment
Python, Linux
The most amazing...
...research I've done was a state-of-art feature selection mechanism—enabling my ML models (alphas) to be considered among the best for out-of-sample performance.
Work Experience
Senior Quantitative Researcher
2022 - PRESENT
Two Sigma Investments
- Developed cryptocurrency market factors and tested their feasibility in explaining risk.
- Researched and developed models for pricing private equity funds.
- Developed a Monte Carlo simulator for portfolio returns with various cashflow options and regime changes.
Technologies: Python, Algorithmic Trading, Data Science, Artificial Intelligence (AI), Machine Learning, Linux, Statistics, Quantitative Research, Financial Forecasting, Research, Fintech
Senior Quantitative Researcher
2022 - 2022
SPS Trading
- Built a backtesting system for analyzing crypto strategies.
- Created and optimized a statistical arbitrage crypto trading strategy using many price-based alphas.
- Developed a clustering algorithm for crypto tokens using the data available on Coin Market Cap.
Technologies: Algorithmic Trading, Python, Genetic Algorithms, Crypto, Data Science, Algorithms, Artificial Intelligence (AI), Machine Learning, Linux, SQL, Quantitative Research, Graphs, Neural Networks, Financial Forecasting, Research, Fintech
Senior Quantitative Researcher
2020 - 2021
Tickup
- Created market-neutral trading strategies on the US equity market and used various datasets to evaluate their performance.
- Focused on NLP techniques for sentiment analysis on news and social media sources to generate trading signals.
- Explored higher latency fundamental trading signals, including predicting revenues using credit card data.
Technologies: Python, SQL, Algorithmic Trading, Artificial Intelligence (AI), Generative Pre-trained Transformers (GPT), Natural Language Processing (NLP), Data Science, Algorithms, Machine Learning, Linux, Quantitative Research, Neural Networks, Financial Forecasting, Research, Fintech
Senior Quantitative Trader
2020 - 2020
Alpha5 Exchange
- Gathered and aggregated order book data from multiple cryptocurrency exchanges.
- Researched high-frequency trading (HFT), market-making strategies for adding liquidity to the exchange, using statistical and ML methods.
- Implemented, debugged, and updated market-making strategies into production.
Technologies: C, Python, Kdb+, Algorithmic Trading, Data Science, Linux, Quantitative Research, Financial Forecasting, Fintech
Quantitative Researcher (Machine Learning)
2018 - 2020
WorldQuant
- Researched trading models on the major equity markets from a large number of datasets, including fundamentals, news, social media, and analyst estimations.
- Handled the full implementation, which included data cleaning, feature engineering, modeling, risk control, and backtesting.
- Named one of the top researchers in 2019 regarding out-of-sample performance in the top 10%.
- Led the team for the ML feature selection project, which developed a more robust state-of-the-art feature selection method for better operating system accuracy.
- Negotiated computational resources and allocation with the GRDs for alpha generated by the method.
- Spearheaded the research efforts into developing sentiment NLP techniques. The resulting trading models were very low correlated and had the largest dollar weight put on by the portfolio managers on average in 2019 out of any other class of models.
- Used different types of neural networks like RNNs, CNNs, and LSTMs on a deep-learning project to predict other targets, experimented with adding macroeconomic features along with the instrument-level predictors.
- Developed batches of signals using genetic algorithms that were updating their parameters live, in production according to their past performance.
- Researched signal data using classification techniques (XGBoost) to help the team produce almost orthogonal strategies to the firm book that helped reduce risk by diversification.
- Composed trading strategies from my signals using portfolio optimization methods with an annualized Sharpe ratio of around 2.5.
Technologies: Python, C++, Algorithmic Trading, Data Science, Algorithms, Artificial Intelligence (AI), Machine Learning, Linux, Statistics, Natural Language Processing (NLP), Quantitative Research, Neural Networks, Financial Forecasting, Numerical Methods, Research, Fintech
Software Engineer
2017 - 2018
ING Bank
- Worked on the core banking algorithms and improved their speed and reliability.
- Analyzed testing information and the core issues of the problems, then designed unit tests for my assumptions and documented solutions for the other developers to implement.
- Tracked and fixed bugs using Jira and the Agile methodology.
Technologies: SQL, Java, Algorithms, Linux, Fintech
Technology Consultant
2016 - 2017
SAP
- Designed, installed, and configured the SAP ERP and database for the client's specific business needs.
- Automated the processes of installation and logging using Bash scripts.
- Implemented custom data reporting and analysis capabilities.
Technologies: Bash, ABAP, Python, SQL, Linux
Experience
Twitter Sentiment Analysis Trading Strategy
As part of a small team, I developed a pipeline for finding the most relevant finance-related Twitter users, scraping their tweets, building sentiment analysis features, and training an ML model for predicting stock returns.
Intraday Trading Platform
I worked on an ongoing Python-based intraday trading platform on cryptocurrency and forex exchanges. This work involved scraping data, building features and models, building a backtesting platform for historical data and paper trading, simulating for slippage and trading costs, integrating exchanges APIs for live trading, optimizing order execution, and developing risk control. I also gathered data from multiple alternative sources like YouTube and Twitter to generate original trading signals.
External PM – Radkl
Live trading a long-short statistical arbitrage crypto strategy as part of an external PM role with the crypto hedge fund Radkl (Dec. 2022 – present).
As of now, the live Sharpe ratio performance is about 2.5.
As of now, the live Sharpe ratio performance is about 2.5.
Oil Market Model Generation and Optimization
Starting from some proprietary data of the contractor on the oil market, I optimized and added to his initial trading signals and created a strategy for daily oil futures trading at a sharp three.
Education
2012 - 2016
Dual Master's & Bachelor's Degrees in Mathematics
University of Cambridge - Cambridge, UK
Certifications
FEBRUARY 2017 - PRESENT
SAP Basis (HANA Database)
SAP
MARCH 2013 - PRESENT
Bronze Medal
South Eastern European Mathematical Olympiad — Greece
MARCH 2012 - PRESENT
Bronze Medal
South Eastern European Mathematical Olympiad — Bulgaria
Skills
Libraries/APIs
Binance API, NumPy, Pandas
Tools
AWS CLI
Languages
Python, C++, R, SQL, ABAP, Bash, Java, C
Paradigms
Quantitative Research, Functional Analysis
Platforms
Linux
Storage
Kdb+
Other
Quantitative Modeling, Data Science, Research, Algorithmic Trading, Machine Learning, Natural Language Processing (NLP), Statistics, Numerical Methods, Financial Forecasting, Genetic Algorithms, Fintech, Quantitative Finance, Sentiment Analysis, Data Analysis, Finance, Artificial Intelligence (AI), Algorithms, Numerical Optimization, Neural Networks, Generative Pre-trained Transformers (GPT), Crypto, Graphs, Epidemiology, Differential Equations, Partial Differential Equations
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