Technical architect · PhD, Computer Science
I do the architecture and the implementation both: schema, ingestion, model, and the screen an analyst actually opens. Government, healthcare and energy.
1B+
clinical records engineered
100M+
public documents ingested
20,000+
patients served
400GB
production database tuned
99.9%
uptime on government infrastructure
€2M+
project value delivered
Domains
Selected work
Healthcare
83,000+ procurement decisions
A self-hosted decision-support system routing between direct retrieval, SQL and chat over a corpus of procurement decisions, with pgvector underneath. Self-hosted because the data could not leave the building.
Government
100M+ documents
Every procurement decision, company registration and public act in the country, ingested from Διαύγεια, ΓΕΜΗ, the Bank of Greece and ΕΣΗΔΗΣ, normalised, and made queryable. Multi-threaded PDF extraction reached a 12× throughput improvement over the sequential design.
Healthcare
100K+ images · 0.89 mAP
A three-tier platform for automated biopsy analysis: an annotation system over 100,000+ medical images, Mask R-CNN tissue segmentation at 0.89 mAP, and a Python inference engine behind a Laravel backend. Published at IEEE BIBM 2023.
Where it comes from
PhD
Computer Science, University of Ioannina. Computer vision and CNN optimisation for medical imaging.
9.0/10
MSc in Computer Science & Networks, with distinction. BSc 8.5/10, with honours.
10
Indexed publications, first author in Elsevier SoftwareX and IEEE BIBM. Publications