Tools & technologies
Natural language processing
Classification, extraction, summarisation and semantic search over text, with transformers and embeddings, from fine-tuning to serving.
Classification, extraction, summarisation and semantic search over text, with transformers and embeddings, from fine-tuning to serving.
From text to a decision
Summarising long documents with chunking, classifying tickets, extracting structured fields or searching by meaning instead of by keyword. Text turned into something actionable.
Understanding the internals
I work with models at a low level when needed: I have implemented a transformer from scratch, attention and tokenizer included, so I know what is happening and where to tune.
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
Transformers and embeddings.
Info graphic
Friction map
Search by meaning.
Info graphic
Journey shape
I know the internals.
Next projects:

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
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
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
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