I currently work at the investments optimization team, aiming at optimizing subsidy in promotions campaigns.- I developed a new time series prediction algorithm that reduced errors in campaign allocations (Prophet, Pandas)- I improved the project ETL so that it became easily extensible to new segmentations (PySpark)

I worked in the Next Best Offer team, which is responsible for recommending subscription offers to Globo’s users. - I developed a new multilabel recommendation algorithm, which achieved a 15% improvement in upsell performance and also reduced training and prediction time by over 80% (ScikitLearn, Pandas) - I worked on maintaining and evolving recommendation algorithms for over 15 million users (ScikitLearn, Pandas, CI/CD, BigQuery, SQL) - I worked on feature extraction from large and complex datasets (Pandas, SQL, BigQuery) - I designed and implemented an A/B test to measure the impact of our solutions (Python, Pandas) - I developed a dashboard to monitor the entire solution (Grafana)

I worked at Globo’s Video Platform aiming at improving users’ Quality of Experience (QoE) with Data Science and Machine Learning.- I implemented Machine Learning models to optimize content caching on Globo’s CDN improving average bitrate- I worked on feature extraction from unstructured data (Pandas, SQL, BigQuery)- I worked with time Series forecasting (Prophet, StatsModels, Tensorflow)- I developed dashboards for models monitoring (Grafana)- I performed contrafactual analysis to measure the impact of our solution (Python, Pandas, Causal Impact)Published Papers:+ "An ML-Based Approach for Near Real-Time Content Caching" (VisNEXT'21, Germany)+ "Aumentando a Eficiência do Cache Proativo com Algoritmos de Mochilas para PoPs e Hashes para Servidores" (WSCad'22, Florianópolis)

Research Subject: Data Science and Machine Learning applied to video and 360-videoPublished Papers:+ "An authoring model for interactive 360 videos" (ICMEW'20, London)+ "A Cluster-Matching-Based Method for Video Face Recognition" (WebMedia'20, São Luís)+ "A Clustering-Based Method for Automatic Educational Video Recommendation Using Deep Face-Features of Lecturers" (ISM'20, Naples)+ "Should i see or should i go: Automatic detection of sensitive media in messaging apps" (WebMedia'21, Belo Horizonte)+ "Video quality enhancement using deep learning-based prediction models for quantized DCT coefficients in MPEG I-frames" (ISM'20, Naples)+ "Decoder-Side Quality Enhancement of JPEG Images Using Deep Learning-Based Prediction Models for Quantized DCT Coefficients" (WebMedia'20, São Luís)+ "Interactive 360-degree Videos in Ginga-NCL Using Head-Mounted-Displays as Second Screen Devices" (WebMedia'20, São Luís)+ "Shaping the Video Conferences of Tomorrow With AI" (WebMedia'20, São Luís)

I worked as Undergraduate Teaching Assistant. My main responsibility was to assist the titular professor and help undergraduate students with their difficulties in Python and C.- Python: lists, recursions, iterations, dictionaries.- C: vectors, pointers, linked lists, recursions, data structures.Titular Professor: Waldemar Celes

- Worked at a consulting project for SulAmerica Auto Seguros in partneship with PUC-Rio. - Worked with mining user profiles and predicting associated risk for them based on driving-related data - Worked with feature enginering with data collected by mobile app Auto.Vc, such as geolocation, circulation hours, driving dynamics, etc.

Data Scientist at the PUC-Rio - Technological Innovation Center Lab in partnership with BTG Pactual. Tools and infrastructure: Python, Scikit-learn, Tensorflow, Keras, Pandas, Numpy, Matplotlib, Linux, Shell Script.Manager: Ruy Luiz Milidiú

Undergratuate researcher at the Nucleus of Applied Computing (NCA) in partnership with Equatorial Energia, which is a holding company with a strong presence in the electricity sector, in the generation, transmission and energy distribution. I worked in the development of an audio recognition algorithm to be used in a mobile application.Manager: Simara Rocha and Aristófanes Corrêa

Research Subject: multimedia authoring, augmented reality, informatics in education and games.Published Papers:+ "BumbAR: Multimedia Authoring Through Augmented Reality" (WebMedia'19, Rio de Janeiro)+ "Exploring an AR-based User Interface for Authoring Multimedia Presentations" (DocEng'18, Halifax)+ "Avaliação da aplicação de jogos no ensino de programação: Uma experiência em uma disciplina introdutória" (SBIE'18)Manager: Carlos de Salles Soares Neto