I build the layer between raw data and decisions: reliable, tested, well-documented data models on top of warehouses like Snowflake and DuckDB, orchestrated with Prefect and served as self-serve analytics. Since January 2025 I've been building analytics infrastructure at RESRG Automotive for 200+ users across 20+ plants.
Static snapshot verified 25 Jul 2026 — dbt test counts read from each project’s manifest.json. Runtime figures fill in once each project’s CI publishes its status.json.
Own the analytics layer of a global operations platform — near-real-time models over 800M+ rows in Snowflake, incremental pipelines, event-driven orchestration, and governed self-serve reporting for 200+ users across 20+ plants.
Research internship — an end-to-end Python framework to evaluate healthcare ML models for fairness, robustness and explainability (XAI).
End-to-end data products I've built. Each has its own page — the business problem, the technical decisions and why, figures read straight from the project's dbt manifest, an honest status section, and what I'd do differently now.
Aggregates EU tech jobs from 5 job-board APIs and answers the one question relocation candidates actually need: will this company sponsor a visa? Every posting's company is matched against the official IND register of recognised sponsors, so each flag carries a KvK number and is auditable — the LLM classifier is only the secondary signal. Runs at €0.
An analytics pipeline that finds under-priced Spanish property listings. Scraped listings flow through validation and idempotent upserts into MotherDuck, a dbt medallion computes an Opportunity Score (€/m² vs. neighborhood benchmarks), and a Streamlit app surfaces the deals.
I'm an Analytics Engineer based in Valencia, Spain. I found my way into data through biomedical engineering, where the most interesting problem wasn't training the model — it was making messy data trustworthy and usable for the people making decisions.
Today I sit between data engineering and analytics: I model data, write tested transformations, orchestrate pipelines, and ship self-serve products. At RESRG Automotive I've built platforms serving 200+ users across 20+ plants in three regions — and on my own time I build ELT pipelines with dbt, DuckDB and Prefect to keep sharpening the craft.
I care about the boring things that make data reliable: clear definitions, tests as contracts, documented lineage, and pipelines that fail loudly instead of quietly lying. I'm now looking for an Analytics Engineer role in a European scale-up, ideally in the Netherlands.
SnowflakeAdvanced
DuckDB / MotherDuck
dbtCore
PythonAdvanced
SQLAdvanced
Prefect
Git & GitHub Actions
Docker
AWS
Power BI
Streamlit
Microsoft Fabric
scikit-learnStart from the decision to be made, not the tool. The stack serves the question.
Data quality isn't a phase. Pipelines should fail loudly, never lie quietly.
Modular, documented, version-controlled models. Lineage anyone can follow.
The best model is the one people actually use. Deliver, measure, refine.
Predictive analytics, multivariate statistics, optimization and statistical process control for industrial applications.
Thesis: "Trustworthy AI Framework for Healthcare" — Grade 10/10
Distinction: ARA UPV — High Academic Performance (Top 25%)
I'm looking for an Analytics Engineer role in a European scale-up. If you're hiring — or just want to talk data modeling, dbt or orchestration — I'd love to hear from you.