Facundo Gilles

Facundo Gilles

Becario Estudiante Avanzado

Followers of Facundo Gilles315 followers
location of Facundo GillesBuenos Aires, Buenos Aires Province, Argentina

Connect with Facundo Gilles to Send Message

Connect

Connect with Facundo Gilles to Send Message

Connect
  • Timeline

  • About me

    Senior Data Scientist @ GlobalLogic | Statistical Learning, Machine Learning Algorithms | PhD Physics

  • Education

    • Universidad Nacional de La Plata

      2012 - 2017
      Doctor of Philosophy - PhD Nanotechnology

      Activities and Societies: The work was carried out at the INIFTA Soft Matter Laboratory facilities, with research stays in the Department of Biomedical Engineering at Northwestern University. Modeling of Physicochemical Properties in ConfinementIn this work, we theoretically/computationally studied the properties of polyelectrolyte layers confined within a nanocanal that traverses a polyethylene terephthalate (PET) membrane from side to side. One of the main contributions of this thesis is the development of a theoretical methodology based on the results of molecular theory for long cylindrical nanochannels.

    • Universidad de Buenos Aires

      2004 - 2012
      Licenciatura Physics

      Effective Interactions in Fluctuating Systems

  • Experience

    • Comisión Nacional de Energía Atómica

      Apr 2009 - Apr 2011
      Becario Estudiante Avanzado

      Scholarship Title: Training for the Pharmaceutical IndustryDuring this scholarship, under the guidance of Dr. Daniel Vega and Dr. Griselda Polla, I acquired hands-on experience in utilizing calorimetric techniques and preparing comprehensive reports specifically tailored to the pharmaceutical industry. The techniques employed encompassed Thermogravimetry (TGA) and Differential Scanning Calorimetry (DSC), enabling me, along with my dedicated research team, to determine critical parameters such as the eutectic point, purity level, and characterization of phase transitions. Together, we conducted parallel research projects, leveraging our collective expertise, which culminated in the publication of a scientific paper in a reputable journal within the field. Show less

    • Universidad de Buenos Aires

      Apr 2009 - Mar 2013
      Ayudante Alumno

      Ayudante Alumno de la Facultad de Ciencias Exactas, cargo ordinario con dedicación simple, rentado.Participe como ayudante en materias de la carrera Lic. en Ciencias Físicas.

    • CONICET

      Apr 2012 - Mar 2017
      PHD Fellow

      Title: Investigation of Physicochemical Properties of Solid-State Nanopores Modified with Polymer MonolayersAbstract: In this thesis, differential equations were derived to describe the thermodynamic state of a complex system confined within solid-state nanopores. Numerical calculation techniques were applied to discretize these equations, and subsequently, high-performance computing (HPC) methods were implemented to solve them. The primary code for solving these equations was developed using Fortran90, while additional scripts in Bash and Python were created to facilitate the management of computational experiments and the analysis of resulting data. The findings of this research were published in reputable journals within the field, contributing to the existing body of knowledge. Show less

    • Universidad Nacional de La Plata

      Apr 2013 - Apr 2017
      Ayudante Diplomado

      Ayudante Diplomado Ordinario con dedicación simple, en las cátedras de Física Experimental I y Física (CIBEX), Taller de enseñanza de la Física.

    • CRISIL Global Research & Analytics

      Apr 2017 - Mar 2019
      Quantitative Analyst
    • Wunderman Thompson Performance Marketing

      Mar 2019 - Dec 2021
      Data Scientist

      As a Data Scientist, I had the opportunity to work on various aspects of the end-to-end development and deployment of both supervised and unsupervised ML models. I gained proficiency in utilizing Python and PySpark to create proof-of-concept models, which helped streamline our processes. I also automated ML models using popular Python libraries such as pandas, numpy, sklearn, xgboost, mllib, papermill, and multiprocessing, among others.In my recent projects, I had the privilege of contributing to the following key initiatives: Audience Segmentation: Objective: To identify, describe, and scale subgroups within a given audience across the full database. Challenge: Developing an unsupervised model (clustering) and determining a suitable metric for selecting the optimal number of segments. Approach: I had the opportunity to build this project from scratch using pyspark, utilizing the k-means approach. To evaluate the stability of the results, we employed normalized mutual information as a metric. Consumer Product Goods: Objective: Scaling knowledge from transactional data to the entire database. Challenge: Effectively preprocessing the data and optimizing the model. Approach: I developed this project from scratch, using an xgboost model. We addressed preprocessing challenges and focused on optimizing the model to ensure accurate predictions. Show less

    • Despegar

      Jan 2022 - Sept 2022
      Senior Data Scientist

      In my role, I held primary responsibility for visitor segmentation and optimizing marketing strategies, with a specific focus on training xgboost models. Additionally, I took the initiative to deliver corporate talks, aimed at introducing the marketing team to the significant potential of applying Machine Learning techniques in the field of marketing. Furthermore, I conducted dedicated sessions aimed at enhancing the team's understanding of Python dataclasses and the fundamental principles of code design. Show less

    • Fourthtech.ia

      Oct 2022 - May 2023
      Senior Data Scientist

      Sports analytics:The primary objective was to implement state-of-the-art models for assessing the performance of soccer players, leveraging thorough research on specialized publications. Furthermore, a significant emphasis was placed on developing customized data visualizations for effectively communicating game insights derived from these advanced models.

    • Universidad Torcuato Di Tella

      Mar 2023 - now

      Computational Methods is an introduction to the world of linear algebra and its applications in computing. Throughout the course, fundamental concepts of matrix algebra, vector space geometry, and optimization methods such as linear programming are explored. The course strikes a balance between theory and practice, providing a solid understanding of key concepts in linear algebra and the necessary tools for programming these techniques in numerical computing frameworks using Python. By the end of the course, students will be prepared to use computational methods in applied problems ranging from engineering and economics to computer graphics and data science. Show less

      • Professor

        Mar 2025 - now
      • Graduate Teaching Assistant

        Mar 2023 - Feb 2025
    • GlobalLogic

      May 2024 - now
      Senior Data Scientist
  • Licenses & Certifications