Gustavo Varela Alvarenga

Regional Head of Data Science for Latin America

About Me

Experienced Head of Data Science with over 20 years of combined academic and professional expertise, specializing in predictive modeling, data product development, and team leadership. Proficient in R, Python, and SQL with substantial experience in deploying data-driven solutions to enhance client decision-making in complex commodity markets. Demonstrates strong capabilities in statistical analysis, machine learning, and data visualization, facilitating effective collaboration across technical and non-technical teams.

Download CV
Interests
  • Time Series Forecasting
  • Artificial Intelligence
  • Machine Learning
  • Applied Statistics
  • Data Science
  • Programming
Education
  • Ph.D. Mathematical Statistics

    University of Maryland

  • MA Economics

    Universidade de Brasilia, Brazil

  • BSc Statistics

    Universidade de Brasilia, Brazil

Experience

  1. Regional Head of Data Science for Latin America

    Argus Media

    Responsibilities include:

    • Established and currently manage the Latin American Data Science team, assembling a group of 12 professionals, including data engineers and scientists across three departments. Two of these departments are led by one department head and one team lead reporting directly to me.
    • Enhanced project delivery efficiency by promoting a culture of innovation and collaboration, resulting in a notable team retention rate of 86% over nearly three years.
    • Spearheaded the creation and deployment of R-based data products, enhancing client decision-making processes and operational efficiency by utilizing automation tools.
    • Cultivated strategic partnerships with shareholders, driving the development of new analytical dashboards and products in R/Shiny, which led to an uplift in stakeholder engagement and product adoption.
    • Collaborated with Business Developers and with the Sales team to design and present data science solutions to clients, enhancing the relevance and impact of presentations, which in turn spiked client interest and engagement.
    • Provided expert guidance in statistical methods and time series analysis, significantly improving project outcomes and team skill levels.
    • In charge of the daily production of the two principal data science products, ensuring the integrity of automated models and the accuracy of results prior to publication.
    • Responsible for reviewing and approving changes to the code-base of the top revenue-generating data science product, instituting robust quality assurance protocols and enforcing strict version control with Git to enhance product reliability and performance.
  2. Lecturer

    University of Maryland

    Responsibilities include:

    • Responsible for the development and delivery of a 400-level course in Computational Statistics, focusing on data wrangling, exploratory data analysis, and inferential statistics (including model building) using SAS© and SQL, directly contributing to the advancement of applied statistics and data science education.
    • Designed and implemented cutting-edge evaluation strategies for a Computational Statistics course, incorporating data-backed research projects that required students to choose a topic, formulate their own research questions, and apply advanced statistical methods such as linear regression, logistic regression, or ANOVA.
    • Emphasized the critical evaluation of methodological assumptions and coding accuracy, significantly enhancing students’ practical skills in data analysis and model interpretation, and preparing them for future careers as data scientists.
  3. National Science Foundation (NSF) Mathematical Sciences Graduate Intern

    Pacific Northwest National Laboratory - PNNL

    Responsibilities include:

    • Completed Coursera’s Deep Learning Specialization, covering from basics to CNNs & RNNs, demonstrating a solid foundation in AI/ML technologies critical for developing advanced data-driven solutions.
    • Led the development and proposal of a RDBMS (PostgreSQL & MySQL) for a DOE, NETL, and PNNL project, showcasing leadership in managing tech projects and contributing to an increase in data processing efficiency by optimizing database architecture.
    • Engineered a Python web crawler to extract relevant data for a PostgreSQL database, and spearheaded the development of a web application for experimental data input, achieving an improvement in data collection accuracy and efficiency.
    • Led a workshop on R programming, focusing on ‘blogdown’ and ‘shiny’ packages, to an all-women audience, promoting diversity in tech and enhancing participants’ data analysis skills.
    Read more about my experience on NSF's website (external link)
  4. Principal Statistician

    Brazilian Agency for Industrial Development (ABDI)

    Responsibilities include:

    • Collaborated with a multidisciplinary team to analyze the impact of public policies (using simulation and quasi-experiment methods -difference in differences, propensity score matching -, and data visualization techniques) on Brazilian industrial development, delivering actionable insights and recommendations. This work was directly showcased through presentations to the Brazilian Minister of Industry.
    • Spearheaded the collection and analysis of diverse socioeconomic data, driving data-driven decision-making and policy formulation. This role highlighted my project management and leadership skills, aligning with the competencies required for a Tech Lead Manager position.
    • Managed data summarization and reporting processes, culminating in strategic recommendations that informed policy improvement initiatives, demonstrating a significant contribution to data-driven policy development.
    • Worked with a team to developed a web enabled tool called The Competitiveness Decoder® that is focused on country data and has the goal of allowing users to learn and master the complexity related to competitiveness through different lenses. Find out more about it here. You can find me under “Former Team Members” on the website.
  5. Lecturer

    Universidade de Brasilia (UnB), Brazil

    Responsibilities include:

    • Transformed the understanding of statistical models among undergraduate students from non-STEM backgrounds, achieving a marked improvement in their ability to apply data insights to diverse fields, as evidenced by enhanced course evaluations and increased student engagement.
    • Pioneered innovative teaching methodologies to demystify complex statistical and machine learning concepts, fostering an interdisciplinary approach to data science education and preparing students for data-driven decision-making roles.
    • Cultivated a collaborative educational environment that bridged the gap between technical and non-technical disciplines, significantly enhancing the interdisciplinary communication skills of students, as measured by their performance in team-based projects.
  6. Statistical Consultant

    Institute for Applied Economic Research (IPEA)

    Responsibilities include:

    • Transformed the understanding of statistical models among undergraduate students from non-STEM backgrounds, achieving a marked improvement in their ability to apply data insights to diverse fields, as evidenced by enhanced course evaluations and increased student engagement.
    • Pioneered innovative teaching methodologies to demystify complex statistical and machine learning concepts, fostering an interdisciplinary approach to data science education and preparing students for data-driven decision-making roles.
    • Cultivated a collaborative educational environment that bridged the gap between technical and non-technical disciplines, significantly enhancing the interdisciplinary communication skills of students, as measured by their performance in team-based projects.
  7. Operational Risk Management - Intern

    Caixa Economica Federal

    Responsibilities include:

    • Wrote advanced SAS macros to automate Value-at-Risk (VaR) calculations and predictions, utilizing convolutions and generalized extreme value distribution (GEV) to significantly enhance risk assessment efficiency by reducing system runtime.
    • Led comprehensive database management and querying initiatives using SAS and SQL procedures, optimizing data retrieval processes and enhancing the accuracy of data-driven decision-making.
  8. Junior Entrepreneur and Marketing Director

    ESTAT Consultoria

    Responsibilities include:

    • Pioneered the university’s first election poll under professorial supervision, establishing a foundation for subsequent academic inquiries.
    • Delivered statistical consulting services to a diverse array of professionals, including sociologists, physicians, and musicians, enhancing their data interpretation and decision-making processes.
    • Designed questionnaires and surveys for market research, focusing on optimizing data quality and reliability.
    • Developed input masks using Microsoft Access and VBA to minimize data collection errors, ensuring higher accuracy in data analysis.
    • Conducted research to refine management and advertising strategies, effectively driving business objectives and creating new opportunities.

Education

  1. Ph.D. Mathematical Statistics

    University of Maryland
    Dissertation on Innovations In Time Series Forecasting: New Validation Procedures to Improve Forecasting Accuracy and A Novel Machine Learning Strategy for Model Selection. Supervised by Prof. Dr. Benjamin Kedem.
    Read dissertation
  2. MA Economics

    Universidade de Brasilia, Brazil
    Thesis on Impacts of the Brazilian science and technology sector funds on industrial firms’ R&D inputs and outputs: new perspectives using a dose-response function. Supervised by Prof. Dr. Donald Pianto. Won the brazilian Premio CNI de Economia in 2012.
    Read thesis
  3. BSc Statistics

    Universidade de Brasilia, Brazil
Skills & Hobbies
Languages
100%
English
100%
Portuguese
Awards and Certifications
Herbert A. Hauptman Fellowship
University of Maryland - Department of Mathematics ∙ December 2021
The fellowship program was created with an estate gift from Carol Fullerton that honors the memory of her late father, Nobel laureate Herbert A. Hauptman (Ph.D. ’55, mathematics), and launched in 2020.
CNI National Economy Award (Prêmio CNI de Economia)
Brazilian National Confederation of Industry (CNI) ∙ December 2012
2nd best paper in the category Evaluation of Productivity and Innovation in the Brazilian Industry. Issued by the Brazilian National Confederation of Industry with the support from the Brazilian Association of Graduate Centers in Economics. Award granted to the paper: ‘‘Impacts of the Brazilian Sector Funds on industrial firms’ R&D inputs and outputs: New perspectives using a dose-response function’’ (original title: Impactos dos Fundos Setoriais nas Empresas: Novas Perspectivas a partir da Função Dose-Resposta). Click here for the full paper in English.
Project Management with Monday.com
LinkedIn ∙ June 2023
I learned how to create boards, add tasks, items, and columns, and invite team members to collaborate effectively. Understood how to manage day-to-day tasks, add updates, and maintain boards by copying, renaming, or deleting them. Discovered how to create folders, additional workspaces, and automations to streamline workflows and increase efficiency.
Neural Networks and Deep Learning
Coursera ∙ July 2019
I studied the foundational concept of neural networks and deep learning. By the end, I was familiar with the significant technological trends driving the rise of deep learning; build, train, and apply fully connected deep neural networks; implement efficient (vectorized) neural networks; identify key parameters in a neural network’s architecture; and apply deep learning to your own applications. Certificate here.
Game Theory
Coursera ∙ May 2012
Completed an in-depth Stanford University course on Game Theory, exploring strategic interactions across varied contexts such as political campaigns and market behaviors. The course involved weekly problem sets, online labs, and a final exam, emphasizing practical applications of game theory through mathematical models. This rigorous coursework honed analytical skills, enhancing abilities in strategic decision-making and predictive analysis.
Model Thinking
Coursera ∙ April 2012
Participated in an advanced course focused on using diverse models to understand complex global phenomena such as political uprisings and market crashes. The course emphasized the strategic application of these models across various fields, including economics, political science, and business, enhancing decision-making and forecasting abilities. Engaged deeply with the material through technical details and algebraic formulations, culminating in successful completion of quizzes and a final exam to earn a Course Certificate. This experience has significantly sharpened my analytical and strategic thinking skills, applicable both academically and in practical scenarios.
Project Management with Monday.com
LinkedIn ∙ June 2023
I learned how to create boards, add tasks, items, and columns, and invite team members to collaborate effectively. Understood how to manage day-to-day tasks, add updates, and maintain boards by copying, renaming, or deleting them. Discovered how to create folders, additional workspaces, and automations to streamline workflows and increase efficiency.
Neural Networks and Deep Learning
Coursera ∙ July 2019
I studied the foundational concept of neural networks and deep learning. By the end, I was familiar with the significant technological trends driving the rise of deep learning; build, train, and apply fully connected deep neural networks; implement efficient (vectorized) neural networks; identify key parameters in a neural network’s architecture; and apply deep learning to your own applications. Certificate here.
Short Courses
Introduction to JavaScript
CodeAcademy ∙ July 2018
Learned JavaScript fundamentals
Learn Python
CodeAcademy ∙ May 2018
Learned Python fundamentals
Topics in Real Analysis
Universidade de Brasilia ∙ February 2012
Summer Course in Real Analysis
Data Visualization with R
Shiny Apps (external page)
Publications
Analytical Skills
  • Impact Evaluation
  • Dataset cleaning
  • Dataset Manipulation
  • Research
  • Econometric Analysis
  • Quantitative and Qualitative Analysis
  • Data Modeling
  • Data Visualization
  • Problem Solving
  • Critical thinking
Statistical and Machine Learning Skills
  • Time Series Analysis
  • Generalized Linear and Mixed Models
  • ANOVA/MANOVA/ANCOVA
  • Instrumental Variables Models
  • Selection Models
  • Matching Algorithms
  • Hypothesis Testing
  • Monte Carlo Simulations
  • Bootstrap Methods
  • Descriptive Statistics
  • Survey design
Dataset Skills
  • Experience with large datasets, including data from household surveys.
  • Had to combine several datasets - with around 50 million lines each and 15 GB each - by an unique identification number.
  • Used SQL procedures in SAS (PROC SQL) for data extraction and manipulation due to the size and complexity of the datasets.