Machine learning, deep learning, neural networks, NLP, data storytelling and data visualisation. Applied projects across all of the above.
Data-driven, AI-powered solutions.
AI pipeline that automated two ends of a heavy documentation cycle: evaluating incoming proposals and assisting the drafting of outgoing specifications.
A unified data platform that broke departmental silos and centralised the organisation’s single source of analytical truth.
Web application that let non-technical users define, maintain, and execute complex operational procedures with full traceability.
Beyond the main case studies, several transversal lines of work in the engineering area:
Deployment of open-source models served with vLLM on a multi-node SLURM cluster, exposed over the internal network to other teams. Included auxiliary services like Marker-PDF on GPU. The organisation gained access to generative AI capabilities without depending on external APIs or exposing sensitive data.
Async platform that lets telescope team leads coordinate and validate usage and performance statistics by cross-referencing data from multiple operational sources. It replaced the dependency on periodic meetings and Excel-style spreadsheets to consolidate scattered metrics, giving each team lead an auditable view they can review on their own schedule.
Work on proofs of concept around emerging technologies. The most recent example: an agentic system that, using natural language and speech-to-text, queries the data it needs from the corporate lakehouse, takes context from a dedicated platform, and generates visualisations on demand that the user can iterate on by simply speaking. The intent is to keep the team in direct contact with what is coming, so adoption decisions are better-informed when the moment arrives.
Contributions to the data platform and AI systems supporting the scientific and administrative operation of the Gran Telescopio Canarias. I work at the boundary between infrastructure, data engineering and on-premise agentic systems.
Applied scientific analysis for municipal services in the aftermath of the 2021 Tajogaite volcano eruption on the island of La Palma. I built the data infrastructure that the emergencies department used for weekly decisions on the affected areas, and delivered findings that changed the post-eruption habitability protocol.
One-to-one tutoring for secondary-school (ESO) and pre-university (Bachillerato) students. In-person and online, with on-demand remote support during exam periods.
Machine learning, deep learning, neural networks, NLP, data storytelling and data visualisation. Applied projects across all of the above.
A strong mathematical base wrapped around the habit of modelling problems and reasoning in the abstract. This was also my first contact with programming, through Python.
Google Data Analytics (Coursera), AWS Cloud Practitioner 101 (AWS Skill Builder), Agentic AI (Udacity), Python and Kubernetes specialisations (Udemy), among others.