Pesquisar

Thiago G.

Python Developer | Data Automation | Machine Learning & Document Processing

(0.0 - 0 avaliações)

Ranking: 2793753 | Projetos concluídos: 0 | Recomendações: 0 | Registrado desde: 04/06/2025

Sobre mim:

Hi! I'm a Python developer with strong expertise in automation, data processing, and machine learning — backed by a formal degree in Mathematics and a postgraduate certificate in Data Science & Analytics.

I develop bots and scripts to extract structured information from PDFs, websites, and spreadsheets. I recently completed my postgraduate program, including a final thesis project where I built a machine learning model for document classification using text mining and NLP techniques. This project is already being used as a product in my current company.

With a solid background in math, statistics, and data workflows, I focus on delivering fast, clean, and scalable solutions to real business problems.

? What I Do:
- Python scripting for automation
- Web scraping (BeautifulSoup, Selenium, Requests)
- PDF parsing and data extraction
- Machine Learning for classification tasks
- Natural Language Processing (NLP)
- Spreadsheet automation (Google Sheets & Excel)
- Large data handling (Pandas, Dask)
- Git version control

? Education:
- B.Sc. in Mathematics
- Postgraduate degree in Data Science & Analytics (USP, Brazil) – Completed

? Strengths:
- Strong analytical thinking
- Clean code & fast delivery
- Clear communication
- Real-world automation & AI experience

Resumo da experiência profissional:

Developed over 115 automation bots to extract structured data from unstructured HTML and PDF fiscal documents (invoices, receipts).
- Delivered JSON-ready outputs used for tax data processing across several Brazilian municipalities.
- Built document layout classifiers using NLP (lemmatization, n-grams, cosine similarity).
- Automated document parsing pipelines using Regex, custom logic, and internal tools.
- Contributed to a lead scoring model using logistic regression with Bayesian adjustment.
- Developed a multi-class document classifier for public procurement documents as part of MBA thesis.

Habilidades:

  • Análise Estatística
  • Aprendizado de Máquina (ML)
  • Modelagem Estatística
  • Python

Áreas de interesse:

  • Outra - Web, Mobile & Software
Carregando...

Carregando...

Pesquisar

FREELANCERS
PROJETOS
Ocorreu um erro inesperado. Caso o erro persista, entre em contato conosco através do e-mail suporte@99freelas.com.br.