José Luis Delgado.
AI and cryptographic systems that can be tested, verified, and audited.
Research engineering for verifiable AI and cryptographic systems.
Current work covers tool-using agents, LLM evaluation, mechanistic interpretability, Lean theorem proving, and post-quantum security. The common thread is reproducible evidence for technical claims.
Research record
José Luis Delgado
Location
Madrid, Spain
jdelgado13@uoc.edu
Research areas
Tool-using agents, LLM evaluation, mechanistic interpretability, Lean theorem proving, post-quantum security
Two lines, one paper index.
Line 01
ML / AI / Formal Systems
Evaluation frameworks, formal reasoning layers, proof-oriented systems, and research workflows with explicit evidence trails.
Line 02
Crypto / PQC / Security
Post-quantum migration, certificate infrastructures, authentication, and operational boundaries in deployed security systems.
Academic background.
2025-2026
M.Sc. in Cybersecurity and Privacy
Universitat Oberta de Catalunya (UOC)
Spain
- Current GPA: 3.8/4. Expected graduation: September 2026.
- Thesis direction: machine-consumable cryptographic governance and post-quantum crypto-agility architectures.
- Advisor: Juan Carlos Fernández Jara.
- Relevant coursework: cybersecurity fundamentals, privacy engineering, blockchain systems, security management, data protection, risk governance, and applied cryptographic infrastructure.
2020-2025
Dual B.Sc. in Software Engineering and Computational Mathematics
Universidad de Tecnología y Arte Digital (U-tad)
Madrid, Spain
- Data Engineering track.
- Bachelor thesis on post-quantum cryptography based on supersingular isogenies, including analysis of the Castryck-Decru attack against SIKE.
- Advisor: Víctor Gayoso Martínez.
- Relevant coursework: machine learning, cryptography and security, algorithms, statistics, optimization, data engineering, software architecture, numerical methods, and discrete mathematics.
Engineering and teaching.
2024-2025
Advanced Analytics and Artificial Intelligence Engineer
Mutua Madrileña
Madrid, Spain
- Built validation checks for production ML workflows, including data-quality tests, feature-store integration, MLflow tracking, and reproducible model documentation.
- Analyzed supervised models with SQL, statistics, and data-quality checks, then turned the results into reports for engineering and business teams.
- Automated repeated review tasks for analytics teams, reducing manual checks and making model results easier to trace back to data and runs.
- Wrote model documentation covering behavior, assumptions, limitations, and operational risks.
2020-2023
Mathematics Teaching Assistant / Tutor
U-tad
Madrid, Spain
- Helped students with calculus, linear algebra, statistics, discrete mathematics, optimization, and formal problem-solving.
- Prepared exercises and review sessions for technical degree students, with an emphasis on clear reasoning and exam readiness.
- Explained abstract definitions, theorems, and symbolic steps through worked examples and practical problem-solving routines.