Who am I?

Current Position: Research Technician in ML for Cancer Research & Materials Simulation (Basque Center for Applied Mathematics, BCAM)

Keywords: Machine Learning \(\cdot\) Bioinformatics \(\cdot\) NLP \(\cdot\) Mathematical Modeling \(\cdot\) Causal ML \(\cdot\) Geometric Deep Learning \(\cdot\) Cancer Research \(\cdot\) Bayesian Modeling \(\cdot\) Hamiltonian Monte-Carlo

Experience

Research Technician in Machine Learning for Cancer Research & Materials Simulation

• Conducted research on how Bayesian approaches combined with Hamiltonian-based Monte-Carlo posterior sampling can be used to model and identify potential genomic biomarkers leading to tamoxifen resistance in breast cancer cells. This was achieved by combining RNA sequencing data from reference cell-lines and patients, with experimental clinical data. This project was done in collaboration with the Cancer Heterogeneity Lab at CIC-bioGUNE.

• Focused on the design & implementation of state-of-the-art adaptive integration schemes within the context of novel Multi-Stage Splitting integrators for Hamiltonian Monte-Carlo.

• Actively developed an open-source project for Hamiltonian-based Monte-Carlo methods: pyHaiCS (Python Hamiltonian for Computational Statistics). An intuitive and efficient library, easy to integrate with existing Python ML pipelines, and built with auto-differentiation, hardware acceleration (such as GPUs and TPUs), parallelization, and pre-compilation of reusable code calls across iterations in mind.

Research Intern in Machine Learning & Artificial Intelligence

• Developed state-of-the-art solutions for noisy artifact removal in Electroencephalogram (EEG) signals based on State Space Models (SSMs) as an alternative to traditional deep sequential models. This work focused on eliminating both ocular and muscular noise, as well as artifacts from transcranial stimulation in studies in the field of Psychology, neuroimaging, and brain stimulation.
• Researched causal geometric deep learning techniques to analyze oppositional narratives in social networks. This involved identifying causal relationships between entities in narrative graphs in order to understand and uncover coordinated attacks in social networks.
• In the process, several research papers as first author have been produced for Q1 journals, as well as some contributions being submitted to international conferences.

Research Technician in Mathematical Modeling with Multidisciplinary Applications

• Conducted an in-depth study into the Effective Sample Size (ESS) metric, a critical tool for evaluating and assessing the quality of Importance Sampling (IS) and Markov Chain Monte Carlo (MCMC) sampling methods. This research was conducted within the framework of defining a new combined metric for accurately assessing the quality of the samples obtained by improved state-of-the-art, Hamiltonian-based joint-MCMC & IS samplers developed at BCAM.
• Proposed an extended ESS metric in the context of IS which incorporates sample-derived corrections by employing sophisticated ML partitioning techniques for the sample space.
• Actively participated in the development, maintenance, and enhancement of the in-house software solution used for these Hamiltonian-based simulations. I collaborated on implementing alternative models, as well as playing a pivotal role in the extensive groundwork required for migrating the software platform from BCAM’s local cluster to the high-performance computing infrastructure at the DIPC supercomputing center.

Education

• MSc in Mathematical Modeling & Research, Statistics, and Computing – University of the Basque Country (UPV-EHU) \(\cdot\) University of Zaragoza
• BSc in Data Science & Artificial Intelligence – University of Deusto

Languages

Skills

• Programming:
  Python, JavaScript, Java, R, Scala, C/C++, LaTeX, HTML, CSS, Shell Scripting, Lua, Swift

• ML Tools:
  Scikit-Learn, TensorFlow, Keras, PyTorch, XGBoost, LightGBM, SHAP, CUDA, JAX

• Databases:
  SQL (MySQL, SQLite, PostgreSQL), Neo4J, MongoDB, SPARQL, Apache Druid

• DevOps:
  AWS, Docker, Git, Apache Superset, Flask, Django, Hadoop, Apache Spark

Publications

Here you can find an updated list of my publications, including journal articles, conference papers, theses, and other contributions.

Conference Papers

A Transformer-Based Approach to Analyzing Public Opinion and Political Trends

Analysing the Impact of Images and Text for Predicting Human Creativity Through Encoders

Theses

A Bayesian Approach to Modeling Tamoxifen Resistance in Breast Cancer Cells Through Adaptive Hamiltonian Monte Carlo Posterior Sampling

Awards & Recognition

• Best Paper Candidate: Analysing the Impact of Images and Text for Predicting Human Creativity Through Encoders – 11th International Conference on Information and Communication Technologies for Ageing Well and e-Health (ICT4AWE), 2025.

Service & Leadership

• Peer-Review Activity: 13 reviews conducted for high-impact journals.