I'm a Data Scientist from Vitória, ES, Brazil 🇧🇷, passionate about leveraging data to build intelligent solutions and drive insights. My core interests lie in Natural Language Processing (NLP), predictive modeling, and developing scalable data pipelines. I enjoy the challenge of translating complex academic research into practical, production-ready systems and have a proven ability to automate workflows for significant efficiency gains.
- Working as a Machine Learning Engineer at Labic , building ETL pipelines to empower the tourism sector with data.
- Contributing as an Undergraduate Researcher at Data Science Lab - UFES, where I've:
- Advancing my knowledge through my B.S. in Computer Science at the Federal University of Espírito Santo (UFES).
Languages:
Machine Learning & NLP: PyTorch | Transformers (BERT) | Scikit-learn | SHAP | FAISS
Cloud, DevOps & Data Tools:
Data Visualization: Matplotlib | Seaborn
You can find more details and other projects on my Portfolio Website
- Episodic Memory for LLMs: Implemented FAISS semantic memory to enhance chatbot coherence.
- Predictive Modeling: Developed high-accuracy churn prediction models.
- Data Engineering: Built ETL pipelines and automated data reporting.
- NLP Research: Processed large text datasets for Portuguese NLP studies.
- LinkedIn: Pedro Igor Gomes de Morais
- Email: pedroigorgm@gmail.com
- Portfolio: https://Pedro2um.github.io/portfolio/
- Kaggle: Pedro2um