# Mariona Carós

Mariona Carós is a remote worker on Himalayas, Data Scientist at Institut Cartogràfic i Geològic de Catalunya, based in Spain.

Source: https://himalayas.app/@marionacars

```text
Data Scientist at Institut Cartogràfic i Geològic de Catalunya building 3D LiDAR deep learning pipelines across Catalonia.
```

| Detail | Value |
| --- | --- |
| Availability | Looking for a job |
| Location | Spain |
| Authorized to work in | Spain |
| Salary expectations | 55k+ EUR |
| Job categories | AI Engineer, Machine Learning Engineer, Deep Learning Engineer, Research Scientist, Lecturer |
| Skills | Machine Learning, Deep Learning, AI, Anomaly Detection, Time series Forecasting, Point Cloud Processing, LiDAR, Computer Vision, Variational Auto Encoders, Convolutional Neural Networks, LSTM, Generative Dialogue Systems, Data Analysis, Statistics |
| LinkedIn | https://linkedin.com/in/marionacaros |

## About

```text
At Institut Cartogràfic i Geològic de Catalunya, I developed deep learning architectures for building change detection and time-series forecasting. I also architected and deployed an end-to-end 3D LiDAR pipeline covering Catalonia.

At Telefónica, I applied machine learning and deep learning to anomaly detection and received an issued patent. Earlier, as a Research Assistant there, I worked on IoT network time-series anomaly detection using a Variational Autoencoder with CNN and LSTM neural networks.

For my research internship at Vodafone, I developed a neural-network-based generative dialogue system for conversational therapy, helping people with mild dementia and early-stage Alzheimer’s exercise memory and communication. My doctoral research at Universitat de Barcelona focused on deep learning models for aerial LiDAR point-cloud data.

I’ve also worked on technology projects with a social impact, including a reminiscence-therapy chatbot and systems for healthcare settings. I teach data science and machine learning at DataScienceUB and instruct deep learning labs at UPC School.
```

## Experience

### Data Scientist (current)

Institut Cartogràfic i Geològic de Catalunya. Jan 2023 - Present

```text
- Developed DL architectures for large-scale building change detection and forecast index time series.
- Architected and deployed an end-to-end 3D LiDAR deep learning pipeline spanning all of Catalonia (32,000 km²), integrating a custom AI model.
- Presented innovation outcomes at international AI forums.
Tools used: VSCode (Python and Pytorch), Jupyter Notebooks, Git, Weights & Biases
```

### Lecturer (current)

DataScienceUB. Jan 2022 - Present

```text
Lecturer at the course "Introducción a la Ciencia de Datos y Machine Learning" of UB
```

### Lecturer (current)

UPC School - Postgraduate Course In Artificial Intelligence With Deep. Feb 2021 - Present

```text
Lab instructor of deep learning models (CNN, GAN, VAE…) and students project advisor
```

### EduCoach (youth mentor)

Fundación "la Caixa". May 2022 - May 2022

```text
Mentor in the fifth and sixth editions of the "EduCaixa Challenge", organized by Fundació La Caixa. Lead 6 teams of national high-school students to develop an initiative related to the United Nations' sustainable development goals through the analysis of data.
```

### Data Scientist

[Telefónica](https://himalayas.app/companies/telefonica). Aug 2019 - Sep 2021

```text
• Application of ML and DL techniques for anomaly detection in several use cases
• Patent issued
• Communication through graphical results to non-technical staff
• Internal talks given on Variational Autoencoders and its applications
Tools used: PyCharm (Python and Pytorch), Jupyter Notebooks, Git
```

### Research Assistant

[Telefónica](https://himalayas.app/companies/telefonica). Mar 2019 - Jul 2019

```text
Anomaly Detection on time series of the 3G and 4G IoT network using a Variational Autoencoder with CNN and LSTM neural networks
```

### Research Internship

[Vodafone](https://himalayas.app/companies/vodafone). Oct 2018 - Mar 2019

```text
Development of a generative dialogue system based on neural networks which consists in a conversational therapy to help patients with mild dementia and early stages of Alzheimer to exercise their memory and communication by evoking past memories.
```

### Internship

everis. Mar 2016 - Dec 2016

```text
Development of a web application using Eclipse in JavaScript, XML and SQL
```

## Education

### Universitat de Barcelona

Doctor of Philosophy, Computer Science. 2021 - 2026

```text
Research and development of deep learning models for 3D point cloud data generated by aerial LiDAR systems provided by the Cartographic Institute of Catalunya.
```

### Coursera

Deep Learning Specialization

```text
Mar 2019
```

### Télécom ParisTech

Bachelor's thesis, Machine Learning for IoT and Anomaly Detection. 2017 - 2017

```text
"Machine Learning with stream processing engines for Internet of Things applications". The Internet of Things (IoT) enables to connect multiple devices for providing a certain service, consequently huge amount of data is generated in time, known as time series. This phenomenon presents unique challenges in defining the data behavior and detecting anomalies.
```

### Universitat Politècnica de Catalunya

Audio-Visual Systems Degree, Telecommunications Engineering. 2012 - 2017

```text
Specialized in image and audio processing
```

### Cambridge English

First Certificate in English (FCE)

### Universitat Politècnica de Catalunya

Master of Engineering - MEng, Telecommunications Engineering. 2017 - 2019

```text
Through this master I obtained skills and expertise in communications systems, networks, electronics and audiovisual systems.
I took the elective subjects on Multimedia based on Deep Learning to specialize in this field.
```

## Tech stack

- PyCharm
- Eclipse
- JavaScript
- Python
- Java
- MATLAB
- XML
- PyTorch
- Git
- SQL
- Weights & Biases

To contact Mariona, message them from this profile with a Himalayas employer account: [Hire on Himalayas](https://himalayas.app/hire).

## People also viewed

- [Cesar Montenegro](https://himalayas.app/@cesarmontenegro): Data Scientist, NLP & LLM, BLUETAB, AN IBM COMPANY
- [Efrén Rama](https://himalayas.app/@efrnrama): AI Engineer, DXC Technology
- [Saurabh B Nagarkar](https://himalayas.app/@saurabhbnagarkar): Research Software Engineer, Barcelona Supercomputing Center
- [Jorge Cayama](https://himalayas.app/@jorgecayama): Data Scientist, SARENET
- [Ignacio Ruiz Chicano](https://himalayas.app/@ignacioruizchicano): Research Fellow, Universidad de Murcia
- [Jose Luis Mellina](https://himalayas.app/@joseluismellina): Predoctoral AI Researcher, University of Murcia
- [Safa Hamreras](https://himalayas.app/@safahamreras): Postdoctoral Researcher in Quantum Inspired Machine Learning, Donostia International Physics Center (DIPC)
- [Vladimir Smirnov](https://himalayas.app/@vladimirsmirnov): AI Engineer, Zynap
- [Manuel Campillo Capel](https://himalayas.app/@manuelcampillocapel): Previsión de la demanda, ElPozo Alimentación
- [Pablo Martinez Seco de Herrera](https://himalayas.app/@pablomartinezsecodeh): Desarrollador IA / ML, Proyecto Fin de Máster
