Data science · ML · Madrid
Data science in Madrid
I work with data to build systems that make decisions: from exploratory analysis and problem definition to a model in production. Tabular data, text and time series, with solid methodology and honest evaluation at every step.
The analysis that precedes the model
Before training anything, I analyse data distributions, missing values, correlations with the target variable and potential biases. A model trained on poorly understood data does not work even if the algorithm is sophisticated.
Feature engineering that contributes
The features you build from raw data usually matter more than the algorithm. I work on feature engineering with domain knowledge: ratios, temporal aggregations, encodings that do not leak information from the future into the past.
Models with methodology
Stratified cross-validation, hyperparameter search with a fixed budget, model comparison on the same test set, probability calibration and feature importance analysis. No shortcuts that inflate results.
Results you can actually use
Notebooks are for exploring, not for delivering. The output of a data science project needs to be a reproducible pipeline, an API or a clear report with actionable conclusions. Not a Jupyter notebook that only runs on my machine.
Have data and want to know what you can do with it?
Write to me with the objective: predict, segment, detect anomalies, understand why something happens. Free first call.
Start a projectFAQ
What is the difference between data science and machine learning?
Data science is broader: it includes exploration, statistical analysis, visualisation and narrative about data. ML is a tool within data science, the part about building predictive models. In practice the two go together in almost every project.
What types of data do you work with?
Mainly tabular data (customer databases, transactions, usage logs), text (emails, tickets, documents, reviews) and time series (sales, consumption, product metrics). For images or audio, consult on the specific scope.
What do I need to start a data science project?
A clear business objective (what decision do you want to make better) and historical data with the variable you want to predict or understand. You do not need perfect data or ML knowledge: the first session is for diagnosing what exists and what is needed.
Can you work with confidential data?
Yes, with appropriate security measures: NDA, work in your environment or a controlled environment, no data on external servers. If the data is very sensitive (health, finance), we discuss it before starting.
How long does a data science project take?
An exploratory analysis with a conclusions report takes 1 to 2 weeks. A complete project (analysis + model + evaluation + delivery) is between 4 and 8 weeks depending on data volume and problem complexity. Deployment and production projects are separate.