The first steps
Security training, then offensive security work at Deloitte from 2019.
Senior AI Security Engineer · Moving into AI Safety
I've worked in security since 2019, first in offensive and application security, then, from 2024, assessing AI systems. Today I'm a Senior AI Security Engineer at CookUnity, building the company's AI Security practice.
Now I'm moving into AI Safety. The problems that pull me in most are misalignment, chain-of-thought monitoring and scheming. I want to help spot when a model pursues goals other than the ones we gave it, and build controls that still hold when it does.
My background is in breaking systems to understand how they fail. That adversarial mindset (assuming something might be actively trying to get past the controls) is what I want to bring to AI safety research while I strengthen my machine learning foundations.
Models that learn goals other than the intended ones.
How useful visible reasoning is for detecting undesired behaviour, and how far it can be trusted.
Models that hide their goals during evaluations.
Cybersecurity → AI security → AI safety
Click, hold and drag to exploreSwipe to explore the journey
In the media
From London to conversations in Bariloche: sharing what I learned at AISB about AI risks and how to contribute to safer development.
The public work I keep going while I build my AI safety foundations: resources, talks and video.
Drafts under review
I am organizing practical approaches to assessing AI systems, from scoping an engagement to documenting findings.
Read the guideAvailable
A local multi-agent tool that turns an architecture's context into a draft threat model to review and challenge. Open source under Apache 2.0.
Explore the projectAvailable
An educational self-assessment to identify gaps and choose what to study next. It is available to use without registration.
Try the assessment
Seven public talks, from Ekoparty to UTN, and a YouTube channel where I document what I learn in AI security.
Remote / Personal publication
United States / Remote
Remote
October 2025 — July 2026
Led the organization's end-to-end AI Security function as the focal point for AI Security within Cybersecurity.
June 2025 — October 2025
Integrated automated security tools (SAST, DAST, SCA) into CI/CD pipelines to expand test coverage and reduce secrets exposure in code repositories.
United States / Remote
Wellington / New Zealand
Buenos Aires / Argentina
Buenos Aires / Argentina
Buenos Aires / Argentina
Co-founder
A cybersecurity community rooted in Argentina and Latin America, developed with the founding team around shared knowledge, events and collaboration.
Ambassador and organizer
I co-founded the Hack The Box meetup in New Zealand and later joined the Argentina chapter as an organizer.
Excerpts from LinkedIn recommendations.
“He was always looking to learn and do his best on every engagement and would actively try to do better on the next one.”“One of his most remarkable qualities is his ability to persist in the face of challenges.”“It is rare to have the opportunity and privilege of witnessing an area being built from the ground up and seeing it grow step by step. That is exactly what I had the chance to experience alongside Ricardo.”If you work on alignment, control or evaluations and think a security background could help, I would like to hear from you. I am also open to AI safety programmes, fellowships and roles.