About me

Email: vituri.vituri@gmail.com
GitHub: github.com/vituri

Professional Summary

Data scientist with a strong background in R, Julia, and mathematics. Experienced in delivering quick data solutions and contributing to open-source and academic communities. Skilled with interactive dashboards in R/Shiny, data pipelines, machine learning models, and scientific research. Publishes on Topological Data Analysis (TDA) and geospatial analytics.

Education

PhD in Topological Data Analysis
Universidade Estadual Paulista (Unesp) / Ohio State University (OSU) 2016 – 2020

• Research on generalizing Vietoris-Rips and Cech constructions using motifs, resulting in new clustering methods for graphs/networks.
  
• Six-month research stay at OSU under Prof. Facundo Mémoli.
  
• Dissertation: Motivic constructions on graphs and networks with stability results.

Master’s Degree – Algebraic Topology
Universidade Estadual Paulista (Unesp) 2014 – 2015

• Studied the Gottlieb group, a subgroup of the fundamental group.
  
• Dissertation: Sobre os grupos de Gottlieb (in Portuguese).

Bachelor’s in Mathematics
Universidade Estadual Paulista (Unesp) 2010 – 2013

Work Experience

Consulting

I offer consulting on R/Shiny, data science, development and machine learning in general. Some projects I’ve worked:

• Geospatial analysis: helping with code and technical writing of books and papers.
• API creation with R and plumber, database modeling, authorization methods and LLMs.
• Dashboard creation using R and Shiny for market research companies.

Senior Data Scientist

Cielo (Remote) September 2025 – Present

Cielo is a major Brazilian financial technology company, specializing in electronic payment processing and providing a diverse range of e-commerce solutions.

• Created and refined machine learning models with Python on big data to predict failures, using tree models and neural networks on noisy time series; presented results to stakeholders with pros-and-cons.
• Automated reports on Databricks to alert clients about needed repairs on their terminals.

R/Shiny Engineer

Appsilon (Remote) Feb 2025 – July 2025

Appsilon provides data science services to the biggest pharma companies in the world, focusing on R/Shiny and quick development.

• Contributed to open source packages and collaborated with Posit to feature applications in the Connect Gallery.
• Delivered a proof-of-concept using AI/LLMs and R/Shiny, automating clinical trial data extraction and reducing processing time from weeks to hours.
  
• Developed and maintained R/Shiny applications using Rhino; fixed issues and reviewed PRs for other open-source packages.
• Contributed to internal knowledge bases and authored blog posts.

Head of Intelligence

Argus Solutions Jan 2020 – Jan 2025

Argus provides solutions for fatigue and distraction detection, control towers, and telemetry in transport operations around the world.

• Founded and led the Data team, growing it to over 12 members, including developers, data scientists, engineers, and mathematicians. Managed hiring, mentoring, and professional development for the team.
• Responsible for creating, testing and deploying 30+ R/Shiny applications for the entire company and clients, used by 300+ different users daily and 20+ users simultaneously. Some of these apps are used by the control tower 24/7 to analyze photos and videos.
• Automated the pipeline of data from hundreds of Excel files to a MariaDB database on AWS RDS, which has 10+ million new entries daily.
• Automated daily/weekly reporting with RMarkdown and Sendgrid, sending 10k+ emails monthly and saving over 20 hours of manual work daily.
• Developed a machine learning model to predict driver drowsiness with >80% accuracy on noisy data, using ensemble methods with tidymodels.
  
• Created a high-performance Julia webserver using Oxygen.jl, reducing data ingestion time from four minutes (with R) to five seconds (with Julia) per iteration.
• Led computer vision projects (cellphone detection, drowsiness, pothole, and gesture detection) using Keras and YOLO in Python.

Technical Skills

• R: Advanced – tidyverse, shiny, geospatial, APIs, package development, optimization
• Julia: Advanced – TDA, performance, APIs, parallelism, documentation
• SQL: Advanced – MariaDB/Postgres, database design, optimization (indexation, normalization), dbplyr within R
• AWS: Intermediate – EC2, S3, RDS, ECS, Docker, Rekognition
• Python: Intermediate – numpy, pandas, polars, scikit-learn
• Databricks: Beginner – delta lake, notebooks, PySpark
• Data Science: Machine learning, computer vision, reporting, dashboards
• Technical Writing: Workshops, blog posts, and academic papers

Selected Projects & Open Source

• JuliaTDA Organization: Owner and main contributor. Wrote the Mapper and Ball Mapper and ToMATo implementations in Julia.
  
• TidierOrg: Contributor to TidierIteration.jl, a Julia version of R’s purrr package.
  
• QuartoDocBuilder.jl: Created a package to facilitate Julia documentation generation with Quarto (docs).
  
• Blog: julia-for-r-users and other posts on R, Julia, and data science. The online version of this CV is here.

Publications, Talks & Workshops

• Paper: Remote sensing to quantify potential aquifer recharge as a complementary tool for groundwater monitoring (Hydrological Sciences Journal, 2024). Co-author; led the R/terra analysis and data pipeline.
• Workshop: Topological Data Analysis workshop (at XXIII Brazilian Topology Meeting, 2024), using Julia.
• Talk: Topology meets the real world: how geometry can help us analyze finite metric spaces (at Workshop of Algebraic Topology and Applications, 2023).
• Paper: Motivic clustering schemes for directed graphs (with Facundo Mémoli, arXiv, 2020).

Additional Information

• Languages: Portuguese (native), English (fluent), Basic Italian and Russian.
• Writing: Author of the forthcoming book, Topological Data Analysis with Julia.
• Interests: R and Julia, open source, topological data analysis, algorithms and performance, technical writing, scientific communication, and mentoring.