Teresa Scholz, Developer in Lisbon, Portugal
Teresa is available for hire
Hire Teresa

Teresa Scholz

Verified Expert  in Engineering

Data Scientist and Developer

Location
Lisbon, Portugal
Toptal Member Since
June 15, 2020

With a Ph.D. in physics, a background in mathematics, and 13 years of experience modeling real-world data, Teresa has the skills to fulfill any data science role. Teresa enjoys working on the whole pipeline from data cleaning and analytics to a final predictive model, especially using machine learning models.

Portfolio

Delivery App Company (for Toptal)
Testing, Model Validation, Pandas, Python 3, Regex...
BNP Paribas
Microsoft Power BI, Tableau, Alteryx, Git, Python
Stratio
Anomaly Detection, Deep Learning, Machine Learning, Keras, Elastic, Jupyter...

Experience

Availability

Part-time

Preferred Environment

Microsoft Power BI, MATLAB, SQL, Git, PyCharm, Keras, Scikit-learn, Pandas, Python

The most amazing...

...thing I've developed was a model to synthesize wind data taking daily wind speed patterns into account in association with a Portuguese wind power producer.

Work Experience

Data Scientist

2020 - 2020
Delivery App Company (for Toptal)
  • Developed a script that cleans the product database using Python and Regular Expressions.
  • Created the ML model that predicts whether or not a product contains alcohol, based on its name.
  • Worked on the development of several models that predict a product's trademark and maker based on its name and description.
  • Revised a model that predicts the dimensions of a product.
Technologies: Testing, Model Validation, Pandas, Python 3, Regex, Natural Language Processing (NLP), GPT, Generative Pre-trained Transformers (GPT)

Data Scientist

2019 - 2020
BNP Paribas
  • Worked on a project that developed an algorithm and a platform to help human resources make hiring choices by predicting future employment and matching candidates to predicted roles.
  • Kick-started and framed a project for a consultancy to match their consultants to available opportunities taking availability and skills in to consideration.
  • Documented a project and identified critical points to be fixed in the next project phase.
Technologies: Microsoft Power BI, Tableau, Alteryx, Git, Python

Data Scientist

2018 - 2019
Stratio
  • Contributed to the development of a method to rate diagnostic trouble codes appearing in the vehicle data which can help fleet management. This work was accepted for publication in the LOD 2020 conference.
  • Researched and developed anomaly detection algorithms (supervised and unsupervised) for various applications to vehicle data using machine learning and deep learning methods.
  • Prepared data, selected features, and engineered using large time-series data.
Technologies: Anomaly Detection, Deep Learning, Machine Learning, Keras, Elastic, Jupyter, Scikit-learn, Pandas, Python

Quantitative Business Analyst

2017 - 2018
Firstwaters
  • Implemented and specified the asset encumbrance report (European banking authority) in Ambit Focus and SQL.
  • Specified and implemented interfaces for regulatory reporting for a newly adapted system. The reporting involved the asset classes ETD, derivatives, securities, and FX/MM.
  • Performed technical and functional testing, defect management, and error analysis.
Technologies: Microsoft Excel, SQL

Ph.D. Candidate

2013 - 2016
University of Lisbon
  • Modeled and analyzed wind turbine data and developed time-independent and time-dependent cyclic Markov models to synthesize data incorporating daily patterns of wind power production.
  • Analyzed the deterministic and stochastic behavior of a wind turbine in the Langevin framework.
  • Developed a parameter-free method to analyze time-series spoiled by strong, correlated measurement noise in the Langevin framework.
Technologies: Stochastic Modeling, Optimization, Python

Research Assistant

2011 - 2013
Laboratorio Nacional de Energia e Geologia
  • Implemented a data-driven segmentation of wind-power time-series using different error measures to identify ramp events.
  • Implemented a partial least squares regression model for the estimation of plasmid, biomass, glucose, glycerol, and acetic acid concentration through FTIR spectra of E. coli bacteria.
  • Implemented a Markov model for a weather pattern time-series resulting from the classification of wind power time-series in terms of atmospheric ciruclation patterns.
Technologies: Scikit-learn, IPython, Python

Prediction Model for the Automotive Industry

I researched and developed an anomaly detection model for a vehicle variable essential in the automotive industry. I researched the topic, collected, and cleaned the data and developed a regression-based anomaly detection using LSTMs.

Libraries/APIs

Matplotlib, NumPy, Pandas, Keras, Scikit-learn

Tools

IPython, Jira, Git, PyCharm, MATLAB, Microsoft Power BI, Jupyter, Elastic, Microsoft Excel, Tableau

Paradigms

Data Science, Anomaly Detection, Testing

Platforms

Jupyter Notebook, Linux, Alteryx

Other

Machine Learning, Statistics, Predictive Analytics, Modeling, Artificial Intelligence (AI), Data Analytics, Research, Mathematics, Data Visualization, Optimization, Neural Networks, Deep Learning, Stochastic Modeling, Natural Language Processing (NLP), Model Validation, GPT, Generative Pre-trained Transformers (GPT)

Languages

Python, SQL, Regex, Python 3

2013 - 2016

Ph.D. in Physics

University of Lisbon - Lisbon, Portugal

2008 - 2010

Master of Science Degree in Medical Technology

Munich University of Technology (TUM) - Munich, Germany

2003 - 2008

Master of Science Degree in Mathematics

Munich University of Technology (TUM) - Munich, Germany

SEPTEMBER 2019 - PRESENT

Sequence Models

Coursera

JULY 2019 - PRESENT

Convolutional Neural Networks

Coursera

MAY 2019 - PRESENT

Improving Deep Neural Networks: Hyperparameter Tuning, Regularization, and Optimization

Coursera

MAY 2019 - PRESENT

Structuring Machine Learning Projects

Coursera

APRIL 2019 - PRESENT

Neural Networks and Deep Learning

Coursera

Collaboration That Works

How to Work with Toptal

Toptal matches you directly with global industry experts from our network in hours—not weeks or months.

1

Share your needs

Discuss your requirements and refine your scope in a call with a Toptal domain expert.
2

Choose your talent

Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.
3

Start your risk-free talent trial

Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.

Top talent is in high demand.

Start hiring