Giulia Falcão

Giulia Falcão

About

Detail

Machine Learning Engineer | AI Engineer | MLOps Engineer
State of Pernambuco, Brazil
Contact Giulia regarding: 
Flexible work
groups
Networking

Timeline

work
Job
school
Education
folder
Project
auto_stories
Publication

Résumé

Jobs verified_user 0% verified
  • ReflexAI
    Senior Machine Learning Engineer public Remote experience
    ReflexAI
    Mar 2025 - Current (1 year 4 months)
    Led and architected the prototype from the ground up, translating requirements into a robust solution by driving system design, performance optimization, cost analysis, and deployment strategies for a MongoDB Atlas Vector Search implementation.
  • Thoughtworks
    Senior Data Scientist public Remote experience
    Thoughtworks
    Dec 2023 - Feb 2025 (1 year 3 months)
    Project 1 - Recommendation system for Pharmaceutical industry Context: Productionized recommendation system using Databricks, Databricks Jobs, Python, Pyspark, MLflow, and AWS S3 leveraging on XGBoost model for better forecasting and planning for a Pharmaceutical Company. Actions: - Developed a fully parameterized Databricks workflow, enabling reusable and scalable pipelines. - Converted code from Databricks notebooks into modular Python scripts to enhance maintainability and collaboration. - Migrated data processing pipelines from Pandas to PySpark, optimizing performance for large-scale datasets. - Analyzed and refactored the client’s data preparation task, implementing spark.cache() to minimize bottlenecks and eliminate redundant com
  • CESAR
    Data Scientist public Remote experience
    CESAR
    Dec 2021 - Nov 2023 (2 years)
    # Project 1 - ETL Pipeline & KPI Dashboard Optimization for a Tech Company Context: Served as a Data Scientist for a prominent American tech client, focusing on projects that involved analyzing key performance indicators (KPIs) by extracting data from MySQL databases. Utilized Python and PySpark to uncover meaningful data patterns and create interactive Tableau dashboards for visualizing KPI findings. Collaborated closely with the Sponsor to define and refine dashboard requirements, ensuring the delivery of accurate and expected results. Developed an ETL process using Python, Docker, and MySQL, optimizing data cleaning, integration, and visualization in Tableau. Actions: - Collaborated on developing the ETL process in Python by defining
  • CESAR
    Junior Data Scientist
    CESAR
    Jun 2021 - Nov 2021 (6 months)
    Project 1 - ETL Pipeline & KPI Dashboard Optimization for a Tech Company Context: Served as a Junior Data Scientist for a prominent American tech client, analyzing key performance indicators (KPIs) by extracting data from MySQL databases. Used Python and MySQL to identify key data patterns and develop interactive Tableau dashboards, delivering accurate and insightful visualizations. Collaborated closely with the Sponsor to refine dashboard requirements and deliver expected results. Developed an ETL process using Python, Docker, and MySQL, optimizing data cleaning, integration, and visualization in Tableau. Actions: - Collaborated on developing the ETL process in Python by defining the architecture using Docker, implementing the pipeline,
  • N
    Machine Learning Engineer
    No Hate AI
    Aug 2020 - Dec 2020 (5 months)
    - Thesis: Developed an automated hate speech detection system, No Hate AI, by applying Natural Language Processing (NLP) techniques to help digital platforms maintain a healthy and non-hostile environment. - Utilized BERT for word embedding generation due to its pre-trained architecture, which performed effectively with limited data, and integrated it into downstream models, including XGBoost, for hate speech classification. - Compared BERT-based embeddings with LSTM-based models to assess performance, ultimately achieving an F1-score of 0.76, outperforming the LSTM-based models with an F1-score of 0.72. - Applied machine learning as a service to deliver scalable and effective solutions, enhancing the accuracy and reliability of hate speec
  • Universidade Católica de Pernambuco
    Undergraduate Student Researcher
    Universidade Católica de Pernambuco
    Aug 2019 - Aug 2020 (1 year 1 month)
    • Developed a non-functional requirements classification system using a Multilayer Perceptron Neural Network (MLP) in Python. • Evaluated the performance of the classification system using F1-Score, Recall, and Acurracy in Python. • Investigated and evaluated how imbalanced the data was and applied an imbalanced algorithm to reevaluate the performance of the classification system.
Education verified_user 0% verified
  • Universidade Católica de Pernambuco
    Bachelor's degree, Computer Science
    Universidade Católica de Pernambuco
    Feb 2016 - Dec 2020 (4 years 11 months)
Projects (professional or personal) verified_user 0% verified
  • F
    Formula 1 Analysis
    Dec 2023 - Current (2 years 7 months)
    Project Requirements (https://github.com/giufalcao/Formula-1/) 1. Data Ingestion Requirements: - Ingest All 8 files into the data lake. - Ingested data must have audit columns. - Ingested data must be stored in columnar format (i.e., Parquet). - Ingested data must have the schema applied. - Must be able to analyze the ingested data via SQL. - Ingestion logic must be able to handle incremental load (Results, PitStopes, LapTimes, Qualifying). 2. Data Transformation Requirements: - Join the key information required for reporting to create a new table. - Transformed tables must have audit columns. - Transformed data must be stored in columnar format (i.e., Parquet). -
Publications verified_user 0% verified
  • X
    Algoritmos de Aprendizagem Supervisionada com Conjuntos de Dados Desbalanceados para Classificação de Requisitos Não-Fun
    XV CONGRESSO BRASILEIRO DE INTELIGÊNCIA COMPUTACIONAL
    Oct 2021