About
I graduated in Physics at the Universitat Autònoma de Barcelona and I'm now doing the Master's in Meteorology at the Universitat de Barcelona.
Most of what I do sits between physical reasoning and code: taking a real dataset, cleaning it until it can be trusted, testing whether a signal is actually there, and turning the result into something other people can explore.
Outside coursework I build things for myself, from a long-term climate analysis of the Vallès Occidental to a training tracker and a Linux server that runs in my home.
Trajectory
From engineering to physics to the atmosphere
2025 — present● in progress
MSc in Meteorology
Universitat de Barcelona (UB)
Atmospheric dynamicsSynoptic meteorologyNumerical weather predictionAtmospheric thermodynamicsSatellite meteorology & remote sensingClimate scienceCompleted July 2025
BSc in Physics
Universitat Autònoma de Barcelona (UAB)
Classical & modern physicsMathematical modellingNumerical methodsStatistical analysisScientific programmingData processing & visualisationBachelor's thesis carried out at ICE-CSIC (Institute of Space Sciences).
Certificate
IBM Data Science Professional Certificate
IBM
PythonData analysisMachine learningBefore Physics
Industrial Engineering (2 years)
Prior studies
Engineering fundamentalsApplied technical disciplines
Direction
Where I want to work
Applied scientific and technical environments, with real physical systems and real datasets rather than purely theoretical research.
- 01
Atmospheric modelling & NWP
How numerical models represent the atmosphere, and how their output is verified against observations.
- 02
Climate data & prediction
Long observational records, trend detection and the statistics needed to say something defensible about change.
- 03
Meteorological applications
Turning environmental data into products people actually use.
- 04
Remote sensing & Earth Observation
Satellite meteorology and gridded observational products.
- 05
Data science & predictive modelling
Statistical analysis and machine learning applied to physical systems, evaluated honestly.
- 06
Wind & solar energy
Resource assessment and forecasting, where meteorology meets engineering decisions.
Toolkit
What I actually use
Only tools that appear in my projects. Where usage is limited, it says so.
Programming & data
07Day-to-day tooling for analysis work.
- PythonAll projects
- NumPy · PandasVO-evolution, thesis
- SciPyVO-evolution trend tests
- MatplotlibVO-evolution, thesis
- Plotly · DashVO-evolution dashboard
- JupyterVO-evolution notebooks
- AstropyThesis (FITS)
Scientific computing
07Methods I have applied to real datasets.
- Time-series analysis
- Trend tests (Mann-Kendall, Sen, Pettitt)
- LOESS smoothing
- Drought indices (SPI / SPEI)
- Data cleaning & QC
- Image fitting (GALFIT)
- Reproducible workflows · pytest
Machine learning
06Used for predictive modelling and evaluated against a physical baseline.
- scikit-learn
- Multilayer Perceptron
- Random Forest
- K-Nearest Neighbours
- XGBoostCoded in the pipeline; not in the final executed comparison
- Cross-validated grid search
Development & infrastructure
06What keeps my own projects running.
- Linux (Debian)
- Git · GitHub
- Command line
- Self-hosted services
- Storage management
- Web development
Approach
How I like to work
Curiosity first
Most projects start with a question I wanted answered, not a tool I wanted to use.
Real systems, real data
I prefer messy observations with gaps and sensor changes over clean toy problems.
Build it to understand it
From a dashboard to a home server, I learn a system by setting it up and keeping it running.
Report what the data supports
A non-significant trend is a result too. I try not to over-read the numbers.