DTx is looking for a researcher to integrate its Data Science and Machine Learning (DSML) Group. The mission of the DSML group? To extract the most value from the data related to the operation of DTx associates. The projects are diverse, but a few recurrent areas of interest can be identified:
- Unsupervised learning
- Anomaly detection and root-cause analysis
- Time-series forecasting
- Explainability of ML algorithms
Domains of application include telecom and logistics (e.g., prevention and detection of fraud or failures), quality control (e.g., maturity estimation of perishable goods, clustering of quality test results from an unlabeled database), business (e.g., cashflow forecasting) and media (e.g., automated highlight detection, recommendation system).
As a researcher in the DSML group, you will:
- Work on one or more projects at a time, in multidisciplinary teams, closely collaborating with DTx associates to address real-world challenges;
- Develop prototypes and proofs-of-concept using state-of-the-art methods in DS and ML;
- Conduct data preparation, exploratory data analysis and feature engineering;
- Find suitable ways to represent data and produce informative visualizations;
- Assess whether to use an ML algorithm for addressing a problem, and choosing the most adequate one(s);
- Run training experiments and hyperparameter optimization;
- Communicate results within the team and to the associates in written and oral forms;
- Produce scientific publications from the results of the projects.
Applicant’s profile and requirements:
- MSc or PhD in a quantitative field: mathematics, statistics, physics, computer science, electronic engineering, computer engineering, or similar;
- Some experience handling data: preparation, exploratory analysis, feature extraction, visualization;
- Some experience with machine learning and knowledge of the basic concepts: classical algorithms, bias and variance, loss functions, gradient descent, regularization, etc.;
- Experience programming in Python, familiar with numpy and pandas;
- Experience using at least one of the following: scikit-learn, Tensorflow/Keras, PyTorch;
- Autonomy and critical thinking.
Extra valuation for expertise in:
- Time-series analysis, in particular for anomaly detection;
- Combinatorial optimization;
- Signal processing (time-frequency analysis, source separation), in particular applied to monitoring of physical assets;
- Autoencoders, self-supervised learning;
- Metalearning, autoML;
- PySpark, DataBricks.
Digital Transformation CoLab DTx carries out its activity of applied research in the areas related to digital transformation.
DTx consists of the following associated members: Accenture; Bosch Car Multimedia; Cachapuz-Bilanciai; CEiiA – Centro de Engenharia e Desenvolvimento de Produto (Centre of Engineering and Product Development); Celoplás; DSTgroup; Aernnova Portugal; IKEA; INL – Laboratório Ibérico Internacional de Nanotecnologia (International Iberian Nanotechnology Laboratory); Mobilleum; NOS; Primavera; Simoldes; TMG Automotive; Universidade Católica Portuguesa; University of Évora; University of Minho; and two affiliated members: CCG – Centro de Computação Gráfica (Center for Computer Graphics) and PIEP – Polo de Inovação em Engenharia de Polímeros (Innovation Centre in Polymer Engineering).
DTx aims at being a reference player in the scope of digital transformation and focuses its research in the intersection of physical, digital and cybernetic domains, with the purpose of creating the next generation of advanced cyber-physical systems.
Application: Submission of the CV and Degree(s) Certificate(s) to the address email@example.com, from January 11th, 2023 to January 25th, 2023, indicating in subject: DTx/72/2023.