Gio’s Data Science is a personal website, designed as a complement to my Curriculum Vitae.

I specialize in Data Science and software development, primarily in Python. Over the years, I’ve had various experiences and different training paths:

  • Studies in humanities and work in the information sector
  • Master’s degree in Banking and Finance (Quantitative Finance path): a highly mathematical-statistical path that involves computer programming for scenario simulation and numerical computation
  • Computer Science studies: a passion for a quarter of a century, over the years I have specialized in the field, developing personal projects and staying updated through manuals, courses, and certifications
  • Currently, I am pursuing a university degree in Computer Science

On these pages, I showcase various activities I have undertaken over the years: side projects, educational activities, and my academic journey.

In my toolkit of knowledge and interests, there is programming on one side and a deep understanding of how financial markets work on the other.

As a Data Scientist and programmer, my focus is on extracting meaning from raw data, so that information becomes useful for answering questions, automating repetitive tasks, making predictions and optimizing processes.

These are craftsman activities that require solving ever-new and different problems by delving into documentation, refining algorithms, and finding new ways to overcome the continuous obstacles that stand between data and the objective. The results can bring significant added value in fields such as finance, energy, industrial processes, marketing or many other sectors where leveraging information and computation is possible.

Typical activities include:

  • Data Extraction
    • Retrieving data from various sources. Searching for datasets, downloading a wide variety of information through REST APIs, web scraping
    • Cleaning and transforming data. Data can have “gaps,”, they may not be intelligible at first sight, or they may have incompatible formats – for example, binary data on legacy systems with special encodings
  • Data Manipulation depending on their type, for example:
    • SQL or various Python libraries to organize, save, separate information from noise
    • Specific functions and mathematical analysis can capture the periodic behavior of time series
    • Specific algorithms can identify relationships in time series with similar trends
    • Handling large datasets – sparse matrices, parallelization
    • GIS: which points of interest, on a map, are within other areas of interest, within a certain radius?
  • Use of Predictive Algorithms, Machine Learning, Artificial Intelligence
    • Clustering, Supervised learning, classification, text mining, time series
    • Scikit-Learn
    • Keras
    • Tensorflow
  • Analysis of Explanatory Variables
    • How much each factor contributes to a certain outcome?
  • Optimization
    • Given a certain process that consumes a certain type of resources, which mix of inputs is worth using to achieve the best economic or performance results?
    • Scipy
    • Tensorflow
  • Simulations
    • Modeling a problem and exploring different scenarios, either through exhaustive execution of all combinations or using the Monte Carlo method
  • Presentation and Interaction
    • Plots
    • Executable applications
    • Web apps
  • Programming: other automatable activities, software development


(Note: I have automatically translated this text)