Tools & technologies
Model evaluation and diagnostics
The part that has taught me the most: reviewing models for data leakage, misleading metrics and optimistic evaluations.
The part that has taught me the most: reviewing models for data leakage, misleading metrics and optimistic evaluations. Starting with my own.
Where it actually breaks
Most models do not fail because of the algorithm, but because of data leakage, a badly chosen metric or an optimistic evaluation. I have learned to look for that first, because it has happened to me.
Judgement, not hype
Sometimes the answer is a simpler model, sometimes it is that you do not need an LLM. I try to keep what actually moves the needle in each case, and several of my open-source tools come from exactly there.
Operating locations
Studying in Madrid
Based in Madrid, learning in public: projects, open-source tools and collaborations, inside and outside Spain.
Info graphic
Decision flow
No hype.
Info graphic
Friction map
Leakage first.
Info graphic
Journey shape
The simple model usually wins.
Next projects:
MLOPSFASTAPIDOCKERMONITORING
Credit Risk Platform
End-to-end MLOps platform for credit-default scoring: reproducible training with a versioned model registry, a FastAPI inference API with input validation, PSI drift monitoring and a Prometheus and Grafana observability stack, all containerised with CI. Code at github.com/delcenjo/credit-risk-platform.
View project
PYTORCHTRANSFORMERSNLP
Transformer from scratch
A GPT-style language model implemented from scratch in PyTorch: multi-head attention, causal masking and residual blocks written by hand, plus a byte-pair tokenizer and an ablation study. Code at github.com/delcenjo/transformer-from-scratch.
View project
NEXT.JSWEBASSEMBLYCLOUDFLARECI/CD
This very site
This portfolio is hand-built with Next.js and deployed on Cloudflare Workers, with a model lab that runs entirely in the browser and continuous deployment on every commit. Code at github.com/jmweb-org/jmweb-pag-web.
View project
SCIKIT-LEARNCALIBRATIONTABULAR
Credit Risk Model
Credit-default risk model with rigorous evaluation: a leakage-free pipeline, cross-validated model comparison, calibrated probabilities, a cost-based decision threshold and per-segment error analysis. Code at github.com/delcenjo/credit-risk.
View project