Professional Experience

Freelance Data Scientist

Freelancer, Dec 2022 - Present

Developed various data science solutions for diverse projects, including building portfolios of financial assets, weather data engineering, and audio data analysis.

  • Technologies: Python, Docker, Econometrics, Machine Learning.

  • Concepts: Data Engineering / Exploration / Analysis / Data Science.

Data Scientist in Quantitative Finance

MBOCHIP, Jun 2021 - Nov 2022

Responsible for analyzing the performance of strategies executed by clients; implementing strategies in Python; maintaining and improving the stock market simulator used for strategy backtesting; analyzing the results of strategy simulations; optimizing strategy hyperparameters; implementing and testing changes in strategies.

Worked directly with optimal execution strategies for large orders such as POV, TWAP, and VWAP, and dealt with both the planning layer of these algorithms and the order execution layer in the stock market.

While developing high-frequency trading strategies, I had the opportunity to work directly with the FIX protocols for receiving market data messages and managing orders from the Brazilian stock exchange, B3.

Developed general-purpose Python packages to solve specific problems faced by the team. These packages facilitated and standardized processes such as: error handling in connections, queries, and insertions in both SQL and NoSQL database services; monitoring the execution of different proprietary applications using logging, and processing market data and trading data in the financial market.

Invested in good practices of clean code, SOLID principles of architecture, and precise and efficient documentation - created with the sphinx library and its extensions - for both the applications themselves and the application development process.

Implemented processes to improve the performance of Python applications, such as: profiling of CPU usage using the line_profiler library; vectorization of calculations and computations using the numpy library; compilation of modules to the C language using the cython library; and parallel execution of processes using the threading and multiprocessing libraries.

  • Technologies: Python (numpy, pandas, statsmodels, matplotlib, psycopg2, pymongo), Docker, PostgreSQL, MongoDB, ElasticSearch, Git and GitLab, Cloud Computing (AWS).

  • Concepts: Quantitative FinanceBig Data, High-Frequency Trading, Market Microstructure, Optimal Execution, Backtesting, Cloud Computing, CI/CD, Software Engineering.

Investment Advisor

Argentum Investimentos, Sep 2018 - Oct 2019

Investment advisory at a licensed XP Investimentos office. Tasks included client prospecting, creation of personalized investment portfolios, monitoring portfolio performance, analyzing client demands, and monitoring investment opportunities available in the market.

Worked with the different asset classes in the Brazilian financial market, such as fixed income securities, stocks and derivatives, investment funds, pension funds, and real estate funds.

Built investment portfolios considering different taxation of financial assets, such as taxation of variable income, fixed income securities, investment funds, and different taxation rules for pension funds.

Advised investors from different suitability classifications, such as qualified and retail investors.

I always considered the specific demands of each client when suggesting portfolios, such as liquidity needs, risk preferences, the dynamics of income and personal expenses of the client, and the client’s short, medium, and long-term personal objectives regarding their wealth.

  • Concepts: Financial Market, National Financial System, Advisory.

Macroeconomic Analyst

Equilíbrio AES, Jan 2017 - Dec 2017

Managed the project ‘Economic Outlook Report’, a quarterly report analyzing the main macroeconomic variables of Brazil (and global variables relevant for the country). The report included an overview of the state of the economy, interpretation of macroeconomic dynamics, and forecast of macroeconomic variables. Statistical and macroeconomic modeling methods were used for the interpretation and forecasting of the analyzed variables. The main work done during the period was the automation of the data collection, analysis, and forecasting process through the creation of a data science application.

Several Brazilian macroeconomic variables were considered, such as GDP, inflation, unemployment, and trade balance. International macroeconomic variables were also taken into account, such as the GDP of Brazil’s main trading partners and the external interest rate.

At the beginning of the year when I contributed as an Analyst, the report had a journalistic character, where data was collected and presented but not interpreted or used for forecasts. The first major change was to make forecasts using econometric models such as the Vector Autoregressive (VAR) model. The model was estimated and forecasts were made using the statistical software Gretl.

Throughout the year, work was done to automate the data collection, cleaning, and processing process using the Python programming language. The purpose of automating data analysis was not only to increase the team’s efficiency but also to create the necessary structure for the use of macroeconomic models for data analysis. With the change, it would also be possible to interpret macroeconomic dynamics. In parallel with the development of the data analysis application, studies of Real Business Cycle (RBC) macroeconomic models were conducted with the aim of using them for the interpretation and forecasting of macroeconomic variables.

  • Technologies: Python (numpy, pandas, statsmodels, matplotlib).

  • Concepts: Macroeconomics, Econometrics, Data Science.