Eduardo J. Barrios García
AI & Software Engineer
Summary
AI Engineer specialising in production-grade RAG systems, LLM infrastructure, and autonomous agents. Sole AI Engineer at one of Spain's largest public institutions, where I outperformed Microsoft Copilot Studio on document retrieval benchmarks. I build systems that move from research to real-world impact — observable, failure-tolerant, and built to iterate.
Technical Skills
Experience
Sole AI Engineer at one of Spain's largest public-sector institutions.
- Designed and shipped a production-grade RAG system handling real government data at scale.
- Outperformed Microsoft Copilot Studio on document retrieval benchmarks using a custom-tuned pipeline on the institution's own corpus.
- Owned every architecture decision independently: retrieval strategy, embedding models, chunking approach, and orchestration layer.
- Led cross-functional coordination — translated institutional requirements into concrete system design across technical and non-technical stakeholders.
Education
Thesis: Applied Multimodal LLMs for Early Glaucoma Detection — Grade: 10/10. Published at XLVI Jornadas de Automática (JA-CEA), Cartagena, Spain.
Publications
Applied Multimodal LLMs for Early Glaucoma Detection Hernández Vidal, J.; Barrios García, E. J.; Alayón Miranda, S.; Díaz Alemán, V. T.
XLVI Jornadas de Automática (JA-CEA), Cartagena, Spain — 2025 · DOI: 10.17979/ja-cea.2025.46.12085