Open to opportunities

João Ferreira

@jooferreira1

Message

Data scientist specialized in maritime emissions, ML pipelines, and scalable data systems.

Portugal

Message

What I'm looking for

I seek roles building production ML and data systems—prefer collaborative teams focused on scalable pipelines, model deployment, and impactful analytics in maritime, sustainability, or computer vision.

I am a data scientist with hands-on experience building end-to-end ML systems, data pipelines, and customer-facing applications that turn maritime AIS data into actionable emissions insights.

At Oceanscore Madeira I developed vessel identification and AIS cleaning algorithms, deployed Django and Streamlit tools, and built prediction models including LSTM approaches for vessel path forecasting.

Previously, I led the first production ML solution at Tracer, scaling image processing with PySpark and PyTorch, integrating generative models for data augmentation, and maintaining automated model retraining and monitoring.

I hold a Master's in Data Science and a Bachelor's in Computing Engineering, and I focus on practical, validated solutions — from optimization-based fuel-mix estimation to deep-learning cargo estimation that materially improved prediction error.

Experience

Work history, roles, and key accomplishments

Current

Data Scientist

Current

Oceanscore Madeira, Lda

Oct 2023 - Present (2 years)

Developed and deployed AIS data-cleaning, vessel identification, and vessel path prediction systems, improving data accuracy for CO2 emissions estimation and enabling scalable emissions reporting for multiple stakeholders. Built deep learning pipelines that reduced cargo estimation MAE from ~12K to ~3K tonnes and delivered customer-facing Streamlit/Django tools for EU ETS and reporting.

Python PyTorch Spark SQL Streamlit Django Databricks Data Cleaning

Data Scientist

Tracer

Oct 2022 - Sep 2023 (11 months)

Led end-to-end development and production deployment of the company’s first ML solution using generative data augmentation and scalable pipelines, processing over 1 billion images and improving model robustness via GANs and Stable Diffusion. Built centralized data repositories and automated model retraining/monitoring.

Pyspark PyTorch Python Snowflake Dataiku Stable Diffusion Data Engineering

Research Assistant

DCC-FCUP

Dec 2021 - Dec 2022 (1 year)

Spearheaded development of an entity linking algorithm for autonomous driving, combining image-derived text with knowledge bases to map entities to domain concepts and iteratively improved performance through evaluation and refinement.

Entity Linking Natural Language Processing Computer Vision Python Knowledge Base Algorithm Development Évaluation

Research Assistant

Interactive Technologies Institute

Sep 2019 - Oct 2020 (1 year 1 month)

Led development of a sustainable development questionnaire platform and implemented the system in Python/Django, conducting an online crowdsourced study to compare household appliance environmental impacts.

Python Django Survey Design Data Collection Crowdsourcing

Education

Degrees, certifications, and relevant coursework

Faculty of Sciences of the University of Porto

Master in Data Science, Data Science

2020 - 2023

Completed a Master in Data Science at the Faculty of Sciences of the University of Porto, covering advanced machine learning, data engineering, and statistical methods.